Geological and geochemical characteristics of the Sawaya`erdun

Transcription

Geological and geochemical characteristics of the Sawaya`erdun
Ore Geology Reviews 32 (2007) 125 – 156
www.elsevier.com/locate/oregeorev
Geological and geochemical characteristics of the Sawaya'erdun
gold deposit, southwestern Chinese Tianshan
Jiajun Liu a,b,c,⁎, Minghua Zheng d , Nigel J. Cook e ,
Xunrong Long d , Jun Deng a , Yusheng Zhai a
a
c
State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences,
Beijing 100083, People's Republic of China
b
Key Laboratory of Ore Deposit Geochemistry, Institute of Geochemistry, Chinese Academy of Sciences,
Guiyang 550002, People's Republic of China
State Key Laboratory of Mineral Deposit Research, Nanjing University, Nanjing 210093, People's Republic of China
d
School of Earth Sciences, Chengdu University of Technology, Chengdu, 610059, People's Republic of China
e
Natural History Museum, University of Oslo, Boks 1172 Blindern,0318 Oslo, Norway
Received 7 October 2004; accepted 12 November 2006
Available online 31 January 2007
Abstract
The Sawaya'erdun gold deposit in the western segment of the South Tianshan, Xinjiang, China, is a large, low-grade orogenic
gold deposit. An abundance of index fossils in the ore-hosting strata, including brachiopoda, corals and Schwagerina, show that the
fossil age of the country rocks ranges from Late Carboniferous to Early Permian. Whole rock Rb–Sr and Sm–Nd isochron ages,
ranging from 292.43 ± 0.38 to 304.7 ± 11.6 Ma, are concordant with a Late Carboniferous age. Gold mineralization is strictly
controlled by a fault zone. Ore minerals are dominated by sulfides, with gold inhomogeneously distributed within them; gangue
minerals are quartz, siderite or calcite. Data suggest that the ore-forming fluid was derived predominantly from an active meteoric
groundwater system. Ore-forming temperatures are estimated to have been within the range 110 to 220 °C.
New 40Ar/39Ar isotope age determinations of auriferous quartz are in excellent agreement with Rb–Sr ages obtained from fluid
inclusions in quartz. They suggest gold mineralization at 206 to 213 Ma (Late Indosinian). Such an age is consistent with the age of
formation given for other gold deposits in the region, such as Bulong and Dashankou. Combined with K–Ar geochronological data
and observed geological features, it is concluded that the Indosinian metallogentic epoch represents a previously unrecognized,
major period of gold metallogeny in the southwestern Chinese Tianshan. In some ways, the Sawaya'erdun gold deposit can be
compared with the Muruntau gold deposit in Uzbekistan with respect to geological setting, host lithology, mineralization style,
mineral assemblages, geochemical association and metallogenic processes involved. Despite this, the Sawaya'erdun gold deposit
possesses a number of characteristics typical of only low-temperature metallogenesis.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Sawaya'erdun gold deposit; Geochronology; Southwestern Chinese Tianshan
Abbreviations: Py = pyrite; Po = pyrrhotite; Hm = hematite; Mt =
magnetite.
⁎ Corresponding author. State Key Laboratory of Geological
Processes and Mineral Resources, China University of Geosciences,
Beijing 100083, People's Republic of China.
E-mail address: liujiajun@cugb.edu.cn (J. Liu).
0169-1368/$ - see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.oregeorev.2006.11.003
1. Introduction
Investigation of the formation conditions of superlarge deposits of gold ore (those in which gold reserves
exceed 1000 t (or 33 Moz), Laznicka, 1983, 1989; Tu,
126
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
2000) and development of exploration models for this
type of deposit are hot topics in modern ore deposit
exploration in China and elsewhere (Tu, 2000; Zhai,
2006). According to statistical data (Liu et al., 1998a),
only nine super-large gold mineral camps have been
discovered throughout the world. One of these is the
Muruntau gold deposit in the Tianshan of Uzbekistan
(Shayakobov et al., 1999) The Tianshan is recognized as
one of the largest gold provinces on Earth and hosts
several world-class gold deposits (Seltmann et al.,
2003). Muruntau, itself, is the largest gold resource in
Eurasia (175 Moz; Goldfarb et al., 2001; Yakubchuk
et al., 2002). Similar, but smaller gold deposits, such as
Amantaitau (4 Moz), Daugyztau (5 Moz), Charmitan
(N10 Moz), Jilau (N 3 Moz), Taror (N 3 Moz), Bakyrchik
(8 Moz), Vasilkovskoye (13.4 Moz), and Kumtor
(18 Moz), are identified in Uzbekistan, Tajikistan,
Kazakhstan and Kyrgyzstan (Goldfarb et al., 2001;
Rui et al., 2002; Yakubchuk et al., 2002), demonstrating
the huge exploration potential for such deposits in the
eastern extension of the Tianshan Mountains in Xinjiang
Province, China (Liu et al., 1999). It is worth noting, at
this point, that Kumtor, another large gold deposit,
occurs in the western extension of Haerketau Mountains, only 40 km west of the frontier with Xinjiang. The
success of exploration in the Central Asian Republics
has provided a great impetus to exploration in the
southwestern Chinese Tianshan Mountains, with participation by a number of multinational exploration companies (Liu et al., 1998a).
The discovery of the Sawaya'erdun gold deposit in
the early 1990s marked an important breakthrough in
prospecting for Muruntau-like orogenic gold deposits
in the Chinese portion of the Tianshan (Zheng et al.,
1996; Li and Luo, 1997; Zheng et al., 1998; Liu et al.,
1998a; Zheng et al., 2001; Liu et al., 2002a). Despite
this, the Chinese mineral exploration community
continues to ask why orogenic Muruntau-like gold
deposits occur so extensively in the southern Tianshan
Mountains yet are largely restricted to the neighboring
countries of China? Secondly, why has it been difficult
to achieve further breakthroughs in prospecting for
gold deposits of this type within the national boundaries of China?
2. Geological setting
The southwestern Chinese Tianshan Mountains are a
portion of the ‘South Tianshan Sb–Hg–Au Metallogenic
Subprovince’. This metallogenic subprovince extends E–
W for more than 3000 km and is separated into western
and eastern parts by the post-mineral Talas-Ferghana
strike-slip fault. The western part of the metallogenic
subprovince is located within the boundaries of the former
Soviet Union, where a large number of orogenic gold
deposits and occurrences have been discovered. The eastern segment (east of Osh) extends into Xinjiang, China.
Great similarities have been recognized in stratigraphy,
petrology and tectonic setting on the two sides of the
border (Zheng et al., 1996; Liu et al., 1998a; Ye et al.,
1998; Liu et al., 2000a; Zheng et al., 2001).
In China, the Tianshan Tectonic Zone can be divided into three belts, i.e., North, Middle and South
Belts, corresponding to the Junggar, North and South
Tianshan Belts on the Central Asian side of the border,
respectively (Liu et al., 1998a). Regional tectonic
lines, such as the Nikolaev fault zone, and ore-hosting
strata, which both control the location of gold deposits,
extend eastwards into the Chinese Tianshan Mountains
(Zheng et al., 1996; Liu et al., 1998a; Zheng et al.,
2001).
The Sawaya'erdun gold deposit, the largest recognized orogenic gold deposit in Xinjiang Province, with
N3 Moz Au proven reserves, and a geologically-inferred
resource of at least 10 Moz Au (Rui et al., 2002), was
discovered by the No.2 Geological Party of Xinjiang
Bureau of Geology and Mineral Resources. The deposit
is located in the western part of Late Paleozoic continental marginal basin of South Tianshan. The northern
part of the deposit continues eastwards to Kyrgyzstan
(Fig. 1). The area around the mine is dominated by strata
of the slightly metamorphosed Upper Carboniferous
clastic flyschoid formation (Zheng et al., 1996; Liu
et al., 1999; Zheng et al., 2001). Carbonate strata,
probably of Middle Devonian age, occur in the southeastern part of the mining district. There is a large, NWdipping reverse fault between the clastic and carbonate
rocks (Fig. 2A).
The ore-hosting clastic rocks can be divided into
lower, middle and upper lithological segments (Fig. 3).
The lower segment is dominated by gray, medium- to
thin-bedded, fine-grained quartz sandstone and siltstone
intercalated with carbonaceous slate. There is thinbedded marl and bioclastic limestone at the base
(Fig. 2B). The middle segment is the most important
ore-hosting part of the succession and is composed of
rhythmically interbedded dark gray to black, thinbedded siltstone and carbonaceous slate. The upper
segment is composed predominantly of gray to dark
gray, medium- to thin-bedded silty slate, siltstone,
pebbled siltstone (Fig. 2C), and carbonaceous slate
with a distinct trend of increasing slate/siltstone ratio
upwards. At the base of the segment is the typical
Bouma sequence (Zheng et al., 1996, 2001).
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Fig. 1. Geological sketch map of the Sawaya'erdun gold deposit (modified after Zheng et al., 1996).
127
128
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
reverse thrust fault in the east and the Yirkshtan thrust
fault in the west. These major structures not only control the distribution of the strata in the region, but they
also play an important role in regional metallogenesis
(Zheng et al., 1996, 2001).
There are no large intrusive massifs in the area.
However, some diorite dykes and diabase dykes in the
west and south of the mining district are positioned
within ore-hosting rocks and underlying strata. All these
dykes strike NNE or close to N–S. In accordance with
their geological and geochemical characteristics, these
dykes are products of at least two different phases of
tectonism (Zheng et al., 2001, 2002; Liu et al., 2002a):
(i) Early Indosinian; K–Ar isotopic ages of the dykes
were 207.5 ± 4.2 to 187 ± 18 Ma; and (ii) Early
Yanshanian, K–Ar isotopic ages of 164.4 ± 2.6 to
137.9 ± 3.8 Ma, which is also consistent with the zircon
U–Pb ages of 133.7 to 131.0 Ma given by Chen and Li
(2003) for monzonite porphyry in this region.
3. Geological characteristics of the deposit
3.1. Ore zones and orebodies
Fig. 2. Photographs of the large fault and fossil sampling locations. A.
The clastic and the carbonate rock strata are separated by a large
reverse fault presented with blue dashed line; B. The Hapsiphyllide,
ostracoda, crinoidal nucha, algae, radiolaria samples were collected
from thin-bedded marl and bioclastic limestone at the bottom of the
lower segment in the ore-hosting strata; C. The Schwagerina samples
were collected from pebbled sandstone and siltstone in the upper
segment of the ore-hosting strata. The areas in the blue outlined box on
the photographs show the expanded images of the arrowed areas.
Abbreviations: CLR = clastic rock; CAR = carbonate rock.
The studied region hosts a number of well developed
faults and folds, extending for several tens to several
hundreds of km from NE to SW. The Sawaya'erdun
gold deposit occurs as a sandwich between the Ariktoru
The Sawaya'erdun gold district covers an area of
about 27 km2 and hosts several tens of fractured zones
of different sizes. The fractured zones are roughly subparallel with the strata (Fig. 1). It has been proven that,
within the mine district, all fractured zones are mineralized without exception. The 24 proven ore zones are
significantly different in size, but the ore zone no. IV is
the largest, with Au reserves exceeding 2.5 Moz (Liu
et al., 1998a). The continuity of mineralization is best
seen in ore zone no. IV, followed by ore zones nos. I, II,
V and VI. The no. IV ore zone is 4200 m long and 15 to
48 m wide. Ore zones nos. I, II and IV are highly
variable in size and are mainly stratoid, lenticular and
vein-like in shape. The main orebody within ore zone
no. IV measures 2870 m in length and up to 48 m in
width over a vertical interval of 300 to 500 m. Typical
gold grades are about 2 g/t, but can be much higher
locally, up to 27 g/t. In addition to Au, grades of Ag, Sb
and W are also relatively high. Silver grades are generally in the range 20 to 70 ppm, with a maximum of
1200 ppm. Tungsten grades attain as much as 150 ppm,
and Sb concentrations can reach as much as 13 wt.%.
The parts of some orebodies are effectively antimony
ores and are located near the hanging wall in the middle
and upper parts of the main orebody, coexisting with Au
orebodies. Antimony orebodies are 180 to 340 m in
length, and 2 to 10 m thick. The typical Sb grades are
1.09 to 2.35%.
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
129
Fig. 3. Stratigraphic column of the ore-hosting clastic rock formation. (1) coral; (2) crinoidal nucha; (3) groove cast; (4) flute cast; (5) coiled bedding;
(6) ripple lamination; (7) parallel bedding; (8) rhythmic stratification; (9) floor underwashing structure; (10) Bouma sequence; (11) graded bedding;
(12) brachiopod; (13) alga; (14) ostracoda; (15) Schwagerina.
3.2. Mineralogy and ore types
3.3. Wall-rock alteration
The deposit is mineralogically complex. Until now,
more than 30 minerals have been identified in the ores,
including native elements, sulfides, sulfates and oxides
of Au, Ag, Fe, Cu, Sb, As, Bi, Pb, Zn, Co, Sn, etc.
(Figs. 4 and 5). The most common ore minerals are
arsenopyrite, pyrite, marcasite, pyrrhotite, chalcopyrite,
jamesonite, stibnite, sphalerite, galena, freibergite,
berthierite, skutterudite, bismuthinite, native bismuth,
electrum and native gold; gangue minerals are mainly
quartz and siderite, with minor calcite and cassiterite.
Owing to intense surface oxidation, secondary oxides are
also extensively developed. The most important are
goethite, lepidocrocite, malachite, anglesite, fibroferrite,
jarosite and scorodite.
According to ore mineral assemblages, primary ores
can be divided into six types: (1) gold-pyrite-quartz ore
(Fig. 6A, B); (2) gold-arsenopyrite-pyrite-quartz ore
(Fig. 6C, D); (3) gold-pyrite-pyrrhotite-quartz ore (Fig.
6E); (4) gold-pyrite-chalcopyrite-sphalerite-galena ore
(Fig. 6F); (5) gold-jamesonite-stibnite ore (Fig. 6G); and
(6) gold-siderite-calcite ore (Fig. 6H).
Wall-rock alteration in the deposit is mainly developed
along and in the vicinity of mineralized fracture zones.
Alteration types include silicification, pyritization, carbonatization, sericitization, and chloritization; the first two
are dominant. Gold mineralization, for the most part, was
emplaced in loci where silicification and pyritization are
most intense. The outstanding feature of silicification is
that the color of the rocks turned from dark- to light-gray,
or even grayish-white, because organic matter was
removed. Sericitization and carbonatization are widely
distributed within carbonaceous slates of the mineralized
fractured zone. Sericite is distributed along schistosity
fissures and coexists with quartz. Sometimes, it is concentrated near quartz veinlets and is hard to distinguish
from sericite formed as a result of regional metamorphism. However, it is relatively coarse and coexists with
other alteration minerals such as pyrite and quartz.
Generally speaking, wherever wall-rock alteration is
intensely developed, sulfides and gold mineralization is
abundant. Where wall-rock alteration is weak, sulfides are
less abundant and gold grades are lower.
130
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Fig. 4. Photomicrographs of mineral assemblages from Sawaya'erdun. All photomicrographs are in reflected light. Diagonal length of view is
0.19 mm. A. Native bismuth is closely associated with pyrrhotite and quartz. Sample S042-1; B. Native bismuth is closely in coexistence with
bismuthinite. Sample S042-1; C. Bismuthinite coexists with galena, quartz and calcite. Sample S036-1; D. Jamesonite coexists with chalcopyrite and
quartz in arsenopyrite. Sample S038-3; E. Jamesonite coexists with chalcopyrite and quartz in pyrite. Sample S039-1; F. Freibergite coexists with
berthierite and pyrite. Sample S040-1; G. Granular cassiterite in pyrite. Sample S042-4; F. Skutterudite in arsenopyrite. Sample S042-5.
Abbreviations: py = pyrite; apy = arsenopyrite; cp = chalcopyrite; po = pyrrhotite; gl = galena; be = berthierite; fg = freibergite; jm = jamesonite; bi =
native bismuth; bis = bismutinite; ct = cassiterite; sk = skutterudite; q =quartz; ca = calcite.
3.4. Ore texture and structure
There exist various ore textures (Fig. 7), including
(1) euhedral, subhedral and anhedral granular and mosaic
textures formed as a result of crystallization; (2) replacement, replacement remnant, pseudomorphous and network textures resulting from metasomatism and filling;
(3) microscopic girdles and colloidal textures formed by
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
131
Fig. 5. Photomicrographs of gold association from Sawaya'erdun. All photomicrographs are in reflected light. Diagonal length of view is 0.19 mm. A.
Electrum from the mineralization stage II (el2, hosted by arsenopyrite) and mineralization stage IV (el4, fracture-hosted by stibnite-quartz veinlet)
Sample S042-2; B. Electrum from the mineralization stage II (el2, hosted by pyrite). Note the sharp boundary between pyrite and arsenopyrite.
Sample S032; C. Electrum from the mineralization stage II (el2, hosted by chalcopyrite). Sample S042-10; D. Electrum from the mineralization stage
III (el3, fracture-hosted by chalcopyrite-quartz veinlet). Sample S041-2; E. Native gold from mineralization stage III, which is closely in coexistence
with pyrite and galena. Sample 99SW-23; F. Electrum from the mineralization stage II (el2, fracture-hosted by chalcopyrite-pyrrhotite-quartz veinlet).
Sample S039-8; G. Electrum from the mineralization stage III (el3, fracture-hosted by chalcopyrite-quartz veinlet). Sample S037-1; H. Electrum from
the mineralization stage II (el2), which is closely in coexistence with arsenopyrite, pyrrhotite, and quartz. Sample 99SW-25. Abbreviations: py =
pyrite; apy = arsenopyrite; cp = chalcopyrite; po = pyrrhotite; gl = galena; stb = stibnite; el = electrum; au = native gold; q = quartz.
colloidal processes; (4) mortar and cataclastic textures
formed under the action of stress. On a larger scale, a
rough zoning with respect to ore structure (brecciation →
fracturing → veining → dissemination) can be recognized
from the center of the orebody outwards. All ore textures
and structure are observed in each ore type.
132
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Fig. 6. Photographs of ore structures and ore types. A. Disseminated ore; B. Massive ore; C and D. Banded ores, with alternating yellow-white pyrite,
dark-grey arsenopyrite and/or white quartz band; E. Quartz-sulfide vein, veinlet and stockwork ore; F. Quartz-cemented breccia vein ore,
disseminated sulfide-bearing. G. Jamesonite vein and brecciated ore; H. Quartz-siderite vein ore, disseminated by pyrite. Abbreviations: py = pyrite;
apy = arsenopyrite; po = pyrrhotite; jm = jamesonite; q = quartz; sid = siderite.
3.5. Gold association
Gold in the ore is present chiefly in the following
forms: firstly as independent minerals (Fig. 5) in primary ores, especially sulfide-rich massive ores, in which
we have found large quantities of electrum and native
gold, coexisting with pyrite, arsenopyrite, chalcopyrite
and quartz. Gold is particularly closely associated with
chalcopyrite. Under the microscope, electrum and native
gold, together with chalcopyrite and quartz, commonly
constitute veinlets and networks penetrating arsenopyrite or pyrite. Occasionally, native gold and electrum
occur independently as micro-fine veinlets (Fig. 5A).
Secondly, gold is mechanically incorporated into host
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
133
Fig. 7. Photographs of ore textures from the Sawaya'erdun gold deposit. All photomicrographs are in reflected light. A. Fracture crosscutting pyrite
contains chalcopyrite and quartz, commonly constituting microfracture network texture. Diagonal length of view is 0.19 mm. Sample S042-10; B.
Fracture crosscutting pyrite contains chalcopyrite and quartz, commonly constituting microfracture network and girdling textures. Diagonal length of
view is 0.38 mm. Sample S041-2; C and D. Arsenopyrite, together with pyrrhotite and pyrite/marcasite, constitute euhedral, subhedral and anhedral
granular textures formed as a result of crystallization. Diagonal length of view is 0.19 mm. Sample S042-1; E. Sphalerite replaces pyrite. Diagonal
length of view is 0.38 mm. Sample S044; F. Chalcopyrite and pyrrhotite replace arsenopyrite and pyrite. Diagonal length of view is 0.38 mm. Sample
S039-2; G. Melnikovite-pyrite and marcasite aggregate, showing colloidal texture. Both formed at equilibrium with respect to each other. Diagonal
length of view is 0.19 mm. Sample S042-2; H. Ring fibre texture of pyrite in quartz. Diagonal length of view is 0.76 mm. Sample S041-2.
Abbreviations: apy = arsenopyrite; py = pyrite; mp y= melnikovite-pyrite; ma = marcasite; cp = chalcopyrite; po = pyrrhotite; sp = sphalerite; au =
native gold; q = quartz; ca = calcite.
minerals. The most important host minerals are pyrite
and arsenopyrite. Gold also can be present in the adsorbed form. Relatively high contents of gold in organic
carbon-rich and oxide ores seem to be related to the
strong capabilities of organic carbon and clay minerals
to adsorb gold.
134
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
3.6. Stages of mineralization
4.1. Collection, pulverization and analysis of samples
Based on the relationships of ore veins, mineral
assemblages, paragenetic sequence (Fig. 8) and ore
fabrics, five stages of mineralization can be recognized:
(1) Stage I, represented by the association of pyritequartz; (2) stage II, represented by the association of
arsenopyrite-pyrite-chalcopyrite-pyrrhotite-electrumquartz; (3) stage III, represented by the common
association of arsenopyrite-pyrite-chalcopyrite-galenasphalerite-stibnite-electrum-gold-quartz; (4) stage IV,
represented by the association of jamesonite-stibnitequartz-gold; and (5) stage V, represented by the
association of pyrite-siderite-calcite. Stages II, III and
IV are the main stages and took place during the
Indosinian epoch at 187∼213 Ma (Liu et al., 2002b,
2004a).
Whole rock samples for Sm–Nd and Rb–Sr
isochronal methods were collected from strata section
a–b and c–d (Fig. 1); the weight of each sample was
generally about 1 kg. In order to take account of samples
altered by late hydrothermal fluids, these samples were
cut and then studied under a microscope. Rock samples
with the same fabric and mineral associations were
chosen and pulverized to b 200 mesh (b0.093 mm); this
material was then analyzed. Both chemical separation
and mass-spectrometric determinations were conducted
at the Yichang Institute of Geology, Chinese Academy
of Geological Sciences and the Modern Analysis Center
of Nanjing University, using high-pressure-cover-melting and positive-ion-exchange techniques with the aid of
MAT-261 and VG-354 isotopic mass spectrometer,
respectively.
The 87Sr/86Sr ratio of American isotopic standard
(NBS987) is determined as 0.710389 ± 9(2σ). The
relative deviation is less than 0.015% compared to the
credential value (0.71034 ± 26). Background values of
Sr and Rb in the whole experiment are 0.20 × 10− 9 g and
0.3 × 10− 9 g, respectively. The 143Nd/144 Nd ratio of the
American La Jolla isotopic standard is 0.511860 ± 6(2σ).
The relative deviation is less than 0.015% compared
with the credential value (0.512633 ± 4). The basal value
of Nd in the whole process is (5–7) × 10− 11 g. The ratios
86
Sr/88Sr = 0.1194 and 146 Nd/144Nd = 0.7219 were used
to calibrate isotopic values of 87Sr/86Sr and 143 Nd/
144
Nd. Analytic errors of 87Sr/86Sr and 143Nd/144Nd
(2σ) were ± 1% and ± 2%, respectively. Rb–Sr and Sm–
Nd isochron ages were calculated using the Ludwig
(1996) ISOTOPE Program (Version 2.90) with 2σ index
for dating error and λRb = 1.42 × 10− 11 a− 1 and λSm =
6.45 × 10− 11 a− 1. The double-error regression method
(York, 1969) is used in calculating Rb–Sr and Sm–Nd
isochron age. Errors on reported ages are given as 2σ.
The fossils were identified by Professors Shi Yan,
Duan Lilan and Gou Zonghai of the Paleontology
Section, Chengdu University of Technology.
4. Age of host rocks
The age of carbonate formation is generally thought
to be Middle Devonian (Zheng et al., 1996; Wang et al.,
2000). No unanimous conclusion can be drawn about
the age of the clastic rock formation, but Li and Luo
(1997), D. Liu et al. (1998a), Ye et al. (1998, 1999a),
and Yang et al. (1999) have proposed a Silurian and
Devonian age. We have undertaken Rb–Sr and Sm–Nd
isotopic analysis and investigation of fossil records in
order to provide new constraints on the age of the orehosting strata.
4.2. Results and discussions
Fig. 8. Paragenetic sequence of main minerals from the five
mineralization stages. Width of lines corresponds to the relative
abundance of minerals of each stage.
4.2.1. Rb–Sr and Sm–Nd isotope ages
Before an abundance of index fossils were found, we
tried to obtain a reliable recognition of the ore-hosting
strata using Rb–Sr and Sm–Nd isotope geochronology.
Five carbonaceous slate samples (S-009, 011, 013, 014,
and 020), five chert samples (98s-20, 31, 33 and 97a-74)
and one sandstone sample (98s-32) were selected.
Results of whole rock Rb–Sr and Sm–Nd isotopic
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
135
Table 1
Results of Rb–Sr isotopic analyses for the whole rock of carbonaceous slates, sandstone and chert from the Sawaya'erdun gold deposit
No.
Sample no.
Lithology
Rb/10− 6
Sr/10− 6
87
87
1
2
3
4
5
6
7
8
9
10
S-009
S-011
S-013
S-014
S-020
98s-20
97a-74
98s-32
98s-33
98s-31
Carbonaceous slate
Carbonaceous slate
Carbonaceous slate
Carbonaceous slate
Carbonaceous slate
Black chert
Gray-green chert
Pyrite-bearing sandstone
Pyrite-bearing chert
Black chert
213.06
140.69
202.71
236.47
296.19
25.35
51.48
63.26
223.4
84.82
131.01
157.03
192.16
178.99
107.39
95.76
25.33
21.29
339.1
79.03
4.6990
2.5868
3.0462
3.8166
7.9811
0.7809
5.5720
6.9980
1.9470
3.1560
0.73104 ± 0.00001
0.72297 ± 0.00001
0.72436 ± 0.00001
0.72882 ± 0.00001
0.74630 ± 0.00001
0.713750 ± 0.000014
0.733705 ± 0.000019
0.739617 ± 0.000022
0.718623 ± 0.000025
0.723134 ± 0.000016
Rb/86Sr
Sr/86Sr ( ± 2σ)
Samples 1–5 samples were analyzed at the Yichang Institute of Geology, CAGS; Samples 6–10 samples were analyzed at the Modern Analysis
Center, Nanjing University.
rock are equal or greater than 650 °C (Dodson, 1976;
Humphrises and Cliff, 1982; Hollister, 1982; Cliff et al.,
1985; Wang et al., 1988; Li, 1996; Fang et al., 2001).
Although ages of carbonaceous slates and sandstones by
Rb–Sr and Sm–Nd isochronal methods probably reflect a
close-to-peak metamorphic age, cherts are more likely to
be reset after closure of their isotopic systems. Generally
speaking, isotopic ages of cherts were not reset during the
later low-grade metamorphism in the area (Knauth and
Epstein, 1976; Jiang, 1989). Therefore, we suggest that
ages determined by the Rb–Sr and Sm–Nd isochronal
method from the chert might have probably represented
the time of sedimentation and/or diagenesis. However,
this age is also consistent with mineralization age at
Muruntau and Kumtor (Mao et al., 2004).
analysis are presented in Tables 1 and 2 and depicted in
Figs. 9 and 10.
In the host rocks of the gold deposit, ratios of 87Rb/86Sr
from carbonaceous slates range from 2.5868 to 7.9811;
ratios of 87Sr/86Sr range from 0.72297 to 0.74630. We
have done a York best-fit analysis of the Rb–Sr isotope
data (Fig. 9A). The calculated slope of this line is
0.004335 ± 0.000166, giving an isochron age of 304.7 ±
11.6 Ma. In contrast, however, 87Rb/86Sr ratios from
cherts and pyrite-bearing sandstones very from 0.7809 to
6.9980; corresponding 87 Sr/ 86 Sr ratios range from
0.713750 to 0.740134. Our York best-fit analysis of this
data (Fig. 9B) gives a calculated slope of 0.00416 ±
0.00005 and an isochron age of 292.43 ± 0.38 Ma.
In the host rocks of the gold deposit, 147 Sm/144 Nd
ratios of cherts and pyrite-bearing sandstones range
from 0.06573 to 0.1621 and 143 Nd/144Nd values range
from 0.511948 to 0.512530. We have done a York bestfit analysis of the Sm–Nd isotope data (Fig. 10). The
calculated slope of this line is 0.001927 ± 0.000125,
giving an isochron age of 294 ± 19 Ma. This age is
similar to that obtained by Rb–Sr methods (Fig. 9).
When attempting to constrain diagenetic or hydrothermal events using isotopic systems, the different isotopic
systems each carry a set of unique implications due to
variation in closure temperatures. Closure temperatures
for Rb–Sr and Sm–Nd isotopic systems for the whole
4.2.2. Fossil age
Since 1995, fossils of algae, radiolaria, brachiopoda,
ostracoda, crinoidal nucha, corals and Schwagerina have
been found in the ore-hosting strata (Liu et al., 1999).
Corals, Schwagerina and brachiopoda (Fig. 11) are the
most useful fossils to provide age data. The coral is
characterized by developed septal fossula and parietes
symmetrical pinnate arrangement (Fig. 11A). It belongs to
Hapsiphyllide Grabau, 1928, and is typical for the Upper
Carboniferous to Lower Permian (Yu et al., 1983). On the
adjoining sagittal section of Schwagerina, an alveolar
Table 2
Results of Sm–Nd isotopic analyses for the whole rock of sandstone and chert from the Sawaya'erdun gold deposit
Serial No.
Sample No.
Lithology
Sm/10− 6
Nd/10− 6
147
143
1
2
3
4
5
98s-20
97a-74
98s-32
98s-33
98s-31
Black chert
Gray-green chert
Pyrite-bearing sandstone
Pyrite-bearing sandstone
Pyrite-bearing sandstone
8.369
3.017
5.028
2.796
9.813
46.82
11.26
26.98
25.74
49.05
0.1077
0.1621
0.1136
0.06573
0.01214
0.512028 ± 0.000008
0.512134 ± 0.000010
0.512039 ± 0.000011
0.511948 ± 0.000007
0.511835 ± 0.000016
Samples were analyzed at the Modern Analysis Center, Nanjing University.
Sm/144Nd
Nd/144Nd ( ± 2σ)
136
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Fig. 9. The whole rock Rb–Sr isochron of the ore-hosting strata from
the Sawaya'erdun gold deposit. A. The Rb–Sr isochronal line of
carbonaceous rocks; B. The Rb–Sr isochronal line of the cherts and
pyrite-bearing sandstones.
texture is observed (Fig. 11B). Although Schwagerina is
widely distributed in the Upper Carboniferous and Lower
Permian, Schwagerinidae is an index fossil for the Upper
Carboniferous (Tan, 1983). The brachiopoda (Fig. 11C)
are small (∼10 to 13 mm ) and are characterized by
circular or pentagonal forms, small and hook umbone,
slight carina of brachial valve, severe convexity of ventral
valve, big notothyrium, narrow notodeltidium, flat dorsal
Fig. 11. The index fossil organisms in the ore-hosting strata. A.
Hapsiphyllide from limestone in the lower ore-hosting strata. Thin
section in plane-polarized light, 3.8 mm diagonal field of view. Sample
S002; B. Schwagerinidae from pebbled sandstone in the upper orehosting strata. Thin section in plane-polarized light, 3.8 mm diagonal
field of view. Sample S079; C. Crurithyris speciosa, Wang and
Crurithyris cf. speciosa, Wang from siltstone in the upper ore-hosting
strata. Sample 99SW-65.
Fig. 10. The whole rock Sm–Nd isochron of the cherts and pyritebearing sandstones from the Sawaya'erdun gold deposit.
face, low or no median fold, shallow or no median sinus,
smooth face or accidentally concentric ornamentation in
hull. They belong to Crurithyris speciosa, Wang and
Crurithyris cf. speciosa, Wang, whose strata range is
Upper Carboniferous to Lower Permian (Tan, 1983).
Because Hapsiphyllide occurs in thinly-bedded marl and
bioclastic limestone at the bottom of the lower segment
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
137
Fig. 12. Chondrite-normalized REE pattern of rocks, ores and hydrothermal minerals in the Sawaya'erdun deposit.
(Fig. 2B), and Schwagerina, brachiopoda occurs in
pebbled sandstone and siltstone from the upper segment
of the ore-hosting strata (Fig. 2C), the fossil age may be
earlier than that of the fossil-bearing strata.
Combining the Rb–Sr and Sm–Nd age data with the
information from the fossil record, we suggest that the
ore-hosting strata probably belong to the Upper Carboniferous rather than to the Silurian, Devonian (Li and
Luo, 1997; Ye et al., 1999a; Wang et al., 2000), or even
Precambrian as had been suggested earlier (Liu et al.,
1998a).
5. Geochemical methodology
Seven hundred fifty two host rock, mineralized rock
and ore samples were collected along sections a–b and
c–d (Fig. 1), from exploration trenches from shallow
pits, drill holes and from the orebodies. Fresh quartz
veins were sampled for 40Ar/39Ar fast neutron activation
and Rb–Sr isochronal analysis from ore zones nos. I and
IV. Firstly, polished thin sections and double-polished
inclusion sections were prepared. These were studied
under the microscope to determine the extent of alteration, to investigate mineral assemblages, ore textures
and inclusion characteristics. In accordance with the
microscopic observations, typical rock, mineralized and
ore samples were selected and ground to − 200 mesh.
The selected ore minerals and gangue minerals were
crushed, sieved and handpicked until grain size reached
40–60 mesh, and the purity exceeded 98%. The ground
quartz sample was then immersed in 5% dilute HNO3
for one hour to remove minor amounts of calcite.
Finally, the pre-treated sample was repeatedly washed in
clean and distilled water and then dried at 80 °C.
Various analyses and measurements were conducted
on the selected samples. For the sake of understanding
138
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
the trace and rare earth element (REE) geochemistry of
the deposit, neutron activation analyses were conducted
on 93 rock, ore and hydrothermally-altered samples.
Trace elements and REE contents were analyzed at the
Laboratory of Neutron Atomic Activation (NAA),
Chengdu University of Science and Technology
(CUST). In order to estimate ore-formation temperature,
the homogenization temperatures and salinities of fluid
inclusions were measured by the senior author at the
Laboratory of Ore Deposits, CUST, and the key
Laboratory of Ore Deposit Geochemistry, Chinese
Academy of Sciences (CAS). In order to constrain the
sources of ore-forming material and age of mineralization, various isotopic measurements were undertaken at
the Stable Isotope Laboratory, Institute of Mineral
Resources, at the Isotope Analysis Center of the
Yichang Institute of Geology and Mineral Resources,
Chinese Academy of Geological Sciences (CAGS), at
the Key Laboratory of Ore Deposits Geochemistry and
at the State Key Laboratory of Environmental Geochemistry, Institute of Geochemistry, CAS, and at the
Isotope Laboratory, Institute of Geology and Geophysics, CAS. Analytical methods are described below.
For S-isotope analyses, Cu2O was used as the oxidizer
for preparation of sulfide samples. Sulfate minerals were
purified to pure BaSO4 by the carbonate-zinc oxide semimelt method, and then SO2 was prepared by the V2O5
oxide method. VCDT standard material was used. For O
and C isotope analysis, calcite and siderite were reacted
with phosphoric acid at 25 °C (McCrea, 1950) to release
CO2. Two Chinese national standards of carbonate for C
and O isotopes, GBW04416 and GBW04417, were used
as working standards in experiments. The δ13CPDB and
δ18OPDB values of GBW04416 determined by NBS-19
were 1.6 and −11.6‰, respectively. The δ13CPDB and
δ18OPDB values of GBW04417 were −6.1 and −24.1‰,
respectively. The δ18OPDB values of calcite and siderite
samples were directly obtained from δ18O values of their
CO2 against the CO2 of PDB. To convert δ18OPDB to
δ18OSMOW, the equation by Friedman and O'Neil (1977)
was used,
d18 OSMOW ¼ 1:03091 d18 OPDB þ 30:91
For oxygen isotope analysis of silica and silicates,
CO2 was prepared by the BrF5 method described by
Fig. 13. Histogram showing homogenization temperature of fluid inclusions in the Sawaya'erdun deposit. A. Mineralization stage I;
B. Mineralization stage II; C. Mineralization stage III; D. Mineralization stage IV; E. Mineralization stage V; F. All mineralization stages.
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
139
Fig. 13 (continued ).
Clayton and Mayeda (1963). For H isotope analysis, the
water in fluid inclusions was released by decrepitation
method. Then the water was reacted with Zn at 400 °C
to produce H2 (Coleman et al., 1982), which was collected in sample tube with active charcoal at liquid N2
temperature.
Silicon was extracted from samples by fluorination
involving reaction with BrF5 at 550 to 600 °C to
produce SiF4 gas (Ding et al., 1994). The SiF4 gas was
purified by liquid N2 and dry ice–acetone separation; it
was then passed through a preheated Cu tube containing
pure Zn particles to remove any remaining BrF5. The
140
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
standard is NBS-28 quartz sandstone provided by the
U.S. National Bureau of Standards and Technology.
Measurements of quartz sandstone standard NBS-28
gave average value of δ28SiNBS-28 = 0.00 ± 0.06‰.
All SO2, SiF4, CO2 and H2 analyses were obtained
with a Finnigan MAT-251 mass spectrometer. Analytical
precision was ± 0.2‰ for C, O, and S isotopes, ± 0.1‰
for Si isotopes, and ± 2‰ for H isotopes.
For Pb isotope analysis, we used the sample pretreatment procedure described by Robin et al. (1999).
Whole-rock samples (400 mg) were dissolved in a mixed
HF, HCl and HBr solution; arsenopyrite, chalcopyrite,
pyrite, pyrrhotite, stibnite and jamesonite (400 mg) were
first washed with HCl to remove impurities coated on the
mineral surfaces, then dissolved in aqua regia under
heating, and finally converted to an HBr chromatographic solution. The Pb in solution was eluted using an
ion exchange resin with HBr–HCl. Of 29 samples, 27
samples were used for analysis on a MAT-261 mass
spectrometer and 2 samples on a VG-354 mass spectrometer. Of the samples, 20% was repeatedly analyzed.
The fractionation factor (0.1% per atomic mass unit) is
worked out in terms of the results of Pb isotopic analyses
of NBS-981 as the international standard, so as to make
correlations for the Pb isotopic ratios of the samples.
Measurements of common-lead standard NBS-981 gave
average values of 206Pb/204Pb = 16.923 ± 3, 207Pb/204Pb =
15.467 ± 3, 208 Pb/204Pb = 36.733 ± 6 for VG-354 and
206
Pb/ 204 Pb = 16.924 ± 3, 207 Pb/ 204 Pb = 15.468 ± 3,
208
Pb/204Pb = 36.732 ± 6 for MAT-261 with uncertainties
b0.1% at the 95% confidence level. The bulk lead blanks
for the whole procedure are b 1 ng and analytical results
for all samples are corrected for this. Duplicate analysis
indicates similar reproducibility in all cases. The bulk
error is b0.1% for lead isotopic samples.
The Rb–Sr mass spectrometric analysis of quartz
fluid inclusions was conducted in accordance with the
procedure of Li et al. (1993). The 87Sr/86Sr ratio determined for NBS987 is 0.710233 ± 54(2σ). The relative
deviation is less than 0.015% as compared with the
credential value (0.71034 ± 26). Background values of
Sr and Rb in the whole experiment are 0.2 × 10− 9 g and
0.3 × 10− 9 g, respectively. The double-error regression
method (York, 1969) is used in calculating Rb–Sr
isochron age on quartz fluid inclusions.
Preparation of quartz samples and standard samples
for analysis, fast neutron irradiation, incremental heating
for Ar extraction and purification, and mass spectrometric analysis were all conducted in accordance with
the procedure of Sang et al. (1994). Monomineralic
samples were wrapped in aluminum foil and placed in
the central part of the B8 hole site of the 49-2 Type
reactor for fast neutron irradiation. Standard samples
used to monitor neutron fluxes include ZBH-25 Biotite,
ZBJ Hornblende from China and GA1550 Biotite from
Australia, whose respective ages are 132.9 ± 1.2 Ma,
132.8 ± 1.4 Ma, and 97.6 ± 0.6 Ma (Sang et al., 1994).
Measurements were conducted on an RGA-10 gas
source mass-spectrometer (VSS Co., U.K.). In static
mode, the blanks of the whole system were determined
as 40Ar = 1.6 × 10− 14 mol and 36Ar = 1.2 × 10− 16 mol.
The decay constant λ is 5.543 × 10− 10/a. Errors involved
in the Ar measurements are within the range 0.5 to 1%.
6. Results
6.1. REE contents and chondrite-normalized REE
patterns
Because the chondrite-normalized REE distribution
patterns of most samples share a number of common
features, the authors selected the NAA results of some
representative samples for discussion. Chondrite-normalized REE distribution patterns are shown in Fig. 12.
Analytical results indicate that ΣREE in sedimentary
and volcanic rocks vary over a relatively large interval
from 2.7 to 328.9 ppm. With the exception of a single
diorite sample, the sample suite is characterized by
LREE/HREE N 4, (La / Yb)N N 4, negative Eu anomalies
(δEu = 0.29 to 0.76) and unremarkable Ce anomalies
(Fig. 12A). The slope for diabase and basalt [(La /
Yb)N N 13] is, in particular, higher than that of the
sedimentary rocks [(La / Yb)N = 4.86 to 8.53]. Total REE
amounts in ores and hydrothermal minerals also are
highly variable (8.49 to 222.1 ppm). In accordance with
their REE distribution patterns, the samples can be
divided into three groups. Samples of the first group
(Fig. 12B) display strong LREE enrichment (LREE/
HREE N 14), with slopes (La / Yb)N of 4 to 8. For this
group of samples, the extent of LREE fractionation
[(La / Sm)N = 3.27 to 5.48] is greater than that of HREE
fractionation [(Tb / Yb)N = 1.3 to 2.38]; negative Eu
anomalies (δEu = 0.37 to 0.84) are prominent. These
features are, in many ways similar to REE distribution
patterns in the sedimentary rocks, but in contrast to the
volcanic rocks, (La / Yb)N exceeds 10. Samples of the
second group (Fig. 12C) show weak LREE enrichment,
with LREE/HREE between 6 and 12, nearly flat REE
distribution pattern curves slope [(La / Yb)N = 1.44 to
3.93], remarkable Eu depletion (δEu = 0.27 to 0.76), Ce
depletion in most cases, and extents of LREE and HREE
fractionation reaching (La/Sm)N = 0.81 to 1.39 and (Tb /
Yb)N = 0.83 to 2.63, respectively. Samples of the third
group are HREE-enriched (Fig. 12D). The samples with
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
such REE characteristics clearly are products of latestage mineralization. Total REE is low (8.49 to
30.07 ppm), with LREE/HREE ratios of 1 to 3. Their
slopes (La / Yb)N are shallow (0.18 to 0.59) and the
extent of LREE fractionation [(Tb / Sm)N = 0.25 to 0.79]
is lower than that of HREE fractionation [(La / Yb)N =
1.17 to 2.46], with no remarkable negative Eu anomaly
(δEu = 0.65 to 0.99).
141
6.2. Fluid inclusion characteristics and estimation of
ore-formation temperature
6.2.1. Fluid inclusion characteristics
Large numbers of fluid inclusions exist in transparent
minerals such as quartz and calcite. General observations include the abundance of fluid inclusions in quartz
relative to calcite, the greater number of highly-
Table 3
δ34S values (‰) of major sulfides from the Sawaya'erdun gold deposit
No.
Sample No.
δ34S− CDT (‰)
Arsenopyrite
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
S042-2
SII-20-2
SII-41
SIV97-1
SIV97-2
S042-1
SI-11
SII-20
SII-20-2
SII-41
SII-42
SII97-3
SIV97-1-1
SIV97-11
SIV97-2
SIV97-23-1
SIV97-3
SIV97-30
SIV97-6
SIV97-22
s0-70-1
s0-70-3-1
S045
SII-19-1
SII-19-2
SIV97-27
SIV97-23-2
SIV97-1-2
s0-70-3-2
SII-19-1
sg23-1
sg-25
sw-12
iv-I-06
sk-01
4-27-02
27kd-02
27kd-03
tc-01
tc24-q1
tc24-q2
Pyrite
Pyrrhotite
Jamesonite
Stibnite
1.01
0.83
0.35
− 0.12
− 0.29
0.23
− 0.09
0.18
0.27
0.16
0.24
0.45
0.65
0.45
0.38
0.26
0.51
2.61
0.71
− 1.35
− 0.28
− 0.19
− 3.00
− 0.77
− 1.02
− 0.48
0.25
0.71
− 0.20
1.10
0.50
− 0.70
− 1.90
− 2.20
0.20
− 0.90
0.20
− 0.20
− 0.90
0.40
− 1.02
Stages of
mineralization
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
Stage
II
III
III
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
III
III
III
IV
IV
IV
IV
II
II
III
IV
I
I
I
III
III
III
III
III
III
III
III
Samples 1–26 were analyzed at the Isotope Analysis Center of the Yichang Institute of Geology and Mineral Resources, CAGS; Samples 27–30 were
analyzed at the Key Laboratory of Ore Deposits Geochemistry, Institute of Geochemistry, CAS; data for the others (31–41) were published by Ye
et al. (1999a).
142
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
transparent quartz inclusions, but smaller number of
poorly-transparent quartz inclusions, that quartz inclusions formed during the main stage of metallogenesis
are more numerous and larger in size than those formed
during the late stage of metallogenesis. Inclusions in
quartz are mostly elliptical, rounded, rectangular or
irregular in shape and vary greatly in size (generally 3 to
15 μm). Inclusions in calcite are mostly rhomb-shaped
and are generally small in size (2 to 6 μm).
Three major types of fluid inclusions have been
distinguished: gas-liquid, liquid, and CO2-rich. Gasliquid and liquid inclusions are all composed of water
and vapor, of which vapor accounts for 3 to 30% of the
total volume. When heating, the vapor phase disappears
and homogenizes to liquid. The CO2-rich three-phase
inclusions are composed of liquid CO2, gaseous CO2
and liquid H2O, of which CO2 accounts for 5 to 7% of
the total inclusion volume. When heated, CO2 disappears in most cases and homogenizes to aqueous
solution; some inclusions decrepitate before reaching
homogenization (Long et al., 1998).
6.2.2. Homogenization temperatures
We systematically measured homogenization temperatures of inclusions in hydrothermal minerals such as quartz
and calcite. Results are 110 to 320 °C for quartz and 100 to
150 °C for calcite. In accordance with the occurrence
characteristics of quartz and calcite, the measured
homogenization temperatures can be divided into five
ranges: 230 to 320, 156 to 233, 138 to 213, 118 to 185, and
96 to 154 °C; Fig. 13), corresponding respectively to the
formation temperatures of quartz and calcite during
mineralization stages I, II, III, IV, and V, respectively.
Except for the ore-barren quartz stage during which
the temperatures were relatively high (220 to 320 °C),
the homogenization temperatures of fluid inclusions in
quartz and calcite associated with the main and late
stages of metallogenesis are all lower than 223 °C, and
the ore-forming temperatures essentially the range between 120 and 210 °C. Ye and Ye (1998) and Ye et al.
(1999a) drew similar conclusions in their studies in that
the deposit appeared to have been formed under epithermal conditions.
6.3. Isotope geochemistry
6.3.1. Sulfur isotopic compositions
Sulfur isotopic analyses were carried out on 41
sulfide samples from different stages of mineralization
(Table 3): 3 pyrite and 1 arsenopyrite samples from
stage I; 15 pyrite and 3 pyrrhotite samples from stage II;
2 arsenopyrite, 11 pyrite and 1 stibnite samples from
stage III; and 5 jamesonite/stibnite samples from stage
IV. Results show that the δ34S values vary from −3.0 to +
2.61‰, averaging 0.05‰ (Fig. 14). The average
measured δ34S values of arsenopyrite, pyrite, pyrrhotite
and jamesonite are 0.73, 0.39, −0.05 and −1.26‰, respectively, following, as expected, the fractionation order of
δ34Sarsenopyrite N δ34Spyrite Nδ34Spyrrhotite Nδ34Sjamesonite. The
sulfides from the different stage have broadly similar δ34S
values with only a narrow range of variation (Fig. 14).
6.3.2. Lead isotopic compositions
As galena is not abundant and is only finely dispersed in the ore, it is very hard to select monomineralic
fractions. The authors therefore carried out Pb isotopic
Fig. 14. Sulfur isotope distributions in the Sawaya'erdun gold deposit.
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
compositional analyses on other sulfides (4 arsenopyrites,
20 pyrites, 3 pyrrhotites, 1 stibnite and 3 jamesonites),
siderite and quartz in the ore and ore-host rocks.
Ore lead isotopic compositions are rather uniform
(Table 4), as evidenced by the Pb isotopic ratios:
206
Pb/204 Pb = 17.965 to 18.348; 207Pb/204Pb = 15.539 to
15.749; and 208Pb/204Pb = 38.062 to 39.029. Pb isotope
compositions are generally consistent with those of
carbonaceous slate and sandy slate, indicating that they
have a common source of lead. However, the singlestage Pb modal ages of ores (181.95 to 472.48 Ma) vary
greatly, reflecting the multi-stage evolutionary character
of the lead (Zheng et al., 2001, 2002; Liu et al., 2002a).
6.3.3. Carbon and oxygen isotopic compositions
Siderite, quartz, and calcite are the most important
gangue minerals in the deposit. The results of C and O
isotope analysis for siderite and calcite in ore and CO2 in
quartz and siderite inclusions are given in Table 5. As can
be seen from the table, δ13C and δ18O values of CO2 in
the fluid are highly variable from −10.50 to +4.98‰
(average on −1.84‰) and +4.4 to 24.2‰ (average on
+ 14.95‰), respectively. As graphite and carbonate
minerals are not seen to coexist in the deposit, the CO2
contents of fluid inclusions are far higher than those of CO
and CH4, i.e., CO2 is the dominant C-bearing constituent
in the hydrothermal system. It can thus be assumed that
δ13Cfluid = δ13CCO2 (Ohmoto, 1972), and therefore, the
δ13C values of the fluid are around −1.84‰. The δ13C
values of CH4 in quartz fluid inclusions vary from −27.7
to −23.9‰, with an average of −25.8‰.
6.3.4. Si isotopic compositions
Silicon isotope analysis was performed on 7 hydrothermal quartz and 6 whole rock samples, the latter
including chert, siltstone, slate and diabase dyke. The
δ30Si values (Table 6) of hydrothermal quartz in the
deposit, except for a single sample that has a relatively
large positive δ30Si value of +0.5‰, all lie within the
range of −0.7 to −0.1‰. Such values are smaller than
those (+0.1 ∼ +1.9‰) of cherts and diabase dyke in the
mine district but are similar to those of ore-hosting
siltstones, fine-grained sandstones and carbonaceous
slates (−0.5 to −0.2‰).
6.3.5. Hydrogen and oxygen isotopic compositions of
fluid inclusions
A total of 25 quartz and siderite samples from different
stage of mineralization were analyzed for δ18O of the host
quartz and siderite, and δD values of fluid from fluid
inclusions. Results are shown in Table 5. The δ18O values
of fluids are calculated from the δ18O of quartz and
143
Table 4
Pb isotopic compositions of the rocks and hydrothermal minerals from
the gold deposit
No. Sample
no.
Mineral and
rock
206
207
208
204
204
204
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Arsenopyrite
Arsenopyrite
Arsenopyrite
Arsenopyrite
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Stibnite
Pyrrhotite
Pyrrhotite
Pyrrhotite
Jamesonite
Jamesonite
Jamesonite
Siderite
Quartz
Slate
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Pyrite
Slate
Slate
Slate
18.116(10)
18.054(10)
17.965(6)
18.062(3)
18.165(10)
18.108(10)
18.259(7)
17.969(8)
18.160(4)
18.102(5)
18.058(4)
18.016(2)
18.088(2)
18.088(2)
18.101(3)
18.348(3)
18.101(2)
18.005(4)
18.062(2)
18.096(1)
18.027(5)
18.081(1)
18.136(2)
18.016(2)
18.002(10)
18.299(10)
18.382(10)
18.099(5)
18.201(12)
18.203
18.123
18.065
18.176
18.012
18.159
18.049
18.144
15.621(10)
15.559(10)
15.553(2)
15.584(3)
15.639(10)
15.638(10)
15.749(8)
15.59(11)
15.582(3)
15.644(6)
15.626(4)
15.557(1)
15.623(1)
15.625(3)
15.627(3)
15.615(3)
15.638(2)
15.539(4)
15.592(3)
15.633(1)
15.573(2)
15.606(1)
15.685(2)
15.576(2)
15.603(10)
15.547(10)
15.637(10)
15.486(16)
15.609(8)
15.580
15.609
15.600
15.570
15.603
15.597
15.584
15.563
38.464(10)
38.210(10)
38.164(5)
38.281(10)
38.597(10)
38.452(10)
38.712(19)
38.239(32)
38.329(9)
38.430(17)
38.355(8)
38.165(3)
38.364(9)
38.385(10)
38.414(7)
39.029(10)
38.431(4)
38.100(8)
38.260(4)
38.399(3)
38.208(8)
38.312(4)
38.579(4)
38.173(16)
38.229(10)
38.062(10)
38.752(10)
38.194(12)
38.581(14)
38.427
38.442
38.311
38.522
38.304
38.228
38.085
38.260
S016-2
S042-2
SII-20-2
SII-41
S016-1
S042-1
s0-70-1
s0-70-3
SI-11
SII-41
SII-42
SII97-3
SIV97-1
SIV97-2
SIV97-23
SIV97-30
SIV97-6
SIV97-22
SIV97-1
SIV97-2
SIV97-23
SII-19-1
SII-19-2
SIV97-27
S039
S026
S020
97sw-23
98sw-33
27kd-03
sk-01
tc24-q2
sg-23-1
sg-25
sg-24
sg-29
sw-11-1
Pb/
Pb(2σ)
Pb/
Pb (2σ)
Pb/
Pb(2σ)
Samples 1–27 were analyzed at the Isotope Analysis Center of the
Yichang Institute of Geology and Mineral Resources, CAGS; Samples
28 and 29 were analyzed at the State Key Laboratory of Environmental
Geochemistry, Institute of Geochemistry, CAS; data for the others
(30–37) were published by Ye et al. (1999a). Data in the brackets are
analytical errors (2σ).
siderite based on the equations 1000lnα(quartz–water) =
3.38 × 106T− 2 − 2.90 (Friedman and O'Neil, 1977) and
1000lnα (siderite–water) = 2.88 × 106T− 2 − 2.70 (O'Neil
and Silberman, 1974) using filling temperatures of fluid
inclusions in quartz (Th). The δD values of fluids were
directly taken from the measurements of from fluid
inclusions. The δ18O and δD values of ore-forming fluids
vary, respectively, from −11.6 to +6.3‰ and from −84 to
−38‰ for different stage of mineralization.
144
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Table 5
The carbon, oxygen and hydrogen isotopic compositions of siderite, quartz and their inclusions
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Sample
no.
Mineral
S010
98j-01
98t-12
98t-16
98t-29
S016
S026
97sw-51
97sw-52
97sw-62
97sw-82
98sm-7
97sw-2
97sw-12
97sw-14
97sw-16
97sw-64
97sw-66
97sw-81
97sw-1
97sw-21
97sw-31
97sw-32
97sw-35
97sw-21
S046
S018
97sw-32
IV-1-9
TC-01
Tc-02
sg-04
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Siderite
Siderite
Siderite
Calcite
Quartz
Quartz
Quartz
Quartz
Stages of
mineralization
Stage I
Stage I
Stage I
Stage I
Stage I
Stage I
Stage II
Stage II
Stage II
Stage II
Stage II
Stage II
Stage III
Stage III
Stage III
Stage III
Stage III
Stage III
Stage III
Stage IV
Stage IV
Stage IV
Stage IV
Stage IV
Stage V
Stage V
Stage V
Stage V
Stage
Stage
Stage
Stage
Fluid inclusions
δ13C
(PDB)
δ18O
(SMOW)
δ13CCO2
(PDB)
− 5.8
− 0.3
− 1.4
− 8.1
− 0.4
− 5.2
− 2.4
δ18OCO2
(SMOW)
δ13CCH4
(PDB)
− 27.7
− 23.7
− 25.6
− 24.5
14.5
16.7
− 3.5
5.0
− 0.3
0.1
0.9
26.0 (25.9)d
− 3.4
− 5.2
18.6
18.0
− 8.7
− 5.3
− 4.5
4.4
− 3.0
− 1.2
2.4
0.0
3.1
5.1
6.2
− 1.6
− 2.0
− 1.5
− 1.8
− 3.3(− 3.6)d
2.9
− 11.6
− 51
− 56
− 80
− 80
− 72
− 59
− 52
− 54
− 61
− 56
− 55
− 43
− 65
− 38
− 50
− 59
− 52
− 65
− 50
− 54
− 69 (− 64) d
− 47
− 68
− 54
− 62
− 6.9
− 6.4
− 6.5
− 7.7
− 65
− 82
− 84
− 67
5.7
2.2
6.3
5.9
4.7
5.2
− 27.4
18.8
17.6
20.5
− 5.3 (−5.4)d
δD b
(SMOW)
24.2
19.0
17.4
19.3
17.1
0.0
− 0.7
0.9
4.8
4.0
− 3.6
− 10.5
δ18OH2O a
(SMOW)
Th c
(°C)
250
272
252
198
205
275
190
192
211
166
205
191
149
183
169
175
153
195
170
124
145
161
153
135
138
125
111
123
Samples Nos. 1–28 were analyzed at the Isotope Analysis Center of the Yichang Institute of Geology and Mineral Resources, CAGS; data for the
others (29–32) were published by Ye et al. (1999a).
a
Calculated based on the equations 1000lnαquartz–water = 3.38 × 106T − 2 − 2.90 (Friedman and O'Neil, 1977) and 1000lnαsiderite–water =
2.88 × 106T − 2 − 2.70 (O'Neil and Silberman, 1974) using filling temperatures of fluid inclusions in quartz (Th).
b
Water from fluid inclusions in quartz, siderite, and calcite.
c
Filling temperatures of fluid inclusions (Zheng et al., 1996; Long et al., 1998).
d
Data in the brackets are repeat analyses.
6.4. Age of mineralization
6.4.1. 40Ar/39Ar isotopic age
Following irradiation, quartz chips (1 to 3 mm in size,
total 60 to 90 mg), having been pre-selected for purity
under a binocular microscope, were analyzed by two bulk
extraction techniques, crushing in vacuum and stepped
heating. Gas released from quartz samples heated in a
tightly closed container was analyzed on a mass
spectrometer for Ar isotopes. The 40Ar/39Ar age data
are listed in Table 7 and results of pre-treatment are shown
in Fig. 15. As can be seen from the figure and table, the
40
Ar/39Ar age spectrum for quartz samples from the
Sawaya'erdun gold deposit is saddle-shaped, which show
that the lowest apparent ages range from 206.09 ± 6.00 to
208.07 ± 1.22 Ma and the plateau ages in the age spectrum
from 208.33 ± 0.55 to 210.59 ± 0.99 Ma.
6.4.2. Rb–Sr isotopic ages on fluid inclusions in quartz
Ye et al. (1999a) and the authors of the present paper
conducted Rb–Sr isotopic analysis on 19 quartz fluid
inclusions from auriferous quartz veins in the
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Table 6
δ30Si values (‰) of the rock and ore-forming quartz veins from the
Sawaya'erdun gold deposit
No.
Sample no.
Mineral and rock
δ30Si− NBS−28
1
2
3
4
5
6
7
8
9
10
11
12
13
97A74
98s-20
97s-23
97A-31
97A59
97A-61
97sw-1
97sw-2
97sw-14
97sw-16
97sw-32
97sw-21
97s-33
Grayish green chert
Black chert
Diabase
Silty sandstone
Siltstone
Carbonaceous slate
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Gold ore
1.9
0.4
0.1
− 0.5
− 0.2
− 0.3
− 0.1
− 0.4
0.5
− 0.1
− 0.1
− 0.5
− 0.7
145
7.2. Genetic implications of stable and radiogenic
isotopes
7.1. Implication of REE distribution patterns
7.2.1. Sulfur isotopes
Sulfides from the different stage all show very narrow ranges of δ34S values. Ore deposits with such sulfur
isotopic characteristics are believed to have formed in a
semi-closed geological environment, or in an environment in which the sulfur was derived from intermediatebasic volcanic rocks in the underlying strata, or, alternatively, sulfur was supplied from a deep source (Chen,
1994; Liu et al., 2000b,c, 2001). Considering the
existence of a number of thick volcanic layers in strata
of various ages underlying the Silurian–Devonian series
in the southwestern Tianshan Mountains, underground
aqueous solutions could extract metals, sulfur and other
components from the volcanic layers when circulating
through them. So these aqueous solutions would be
important suppliers of sulfur and other components. The
data, with δ34S values for different minerals following
the anticipated order, indicate consistency with sulfur
isotope equilibrium fractionation.
According to Ohmoto (1972), S-bearing mineral assemblages are formed from hydrothermal solutions with
certain ΣS isotopic composition and its δ34S values
vary with T, pH, ƒO2, etc. Particularly when log ƒO2 is
relatively high (N− 42), δ34S values are very sensitive to
the variation in ƒO2 and pH. Slight variations in ƒO2
and pH will lead to a great variation in δ34S (Liu et al.,
2000b). With the evolution of hydrothermal activity, if
the early minerals were enriched in 34S, the residual
hydrothermal solutions would be relatively enriched in
32
S. Therefore, minerals crystallized from the residual
solutions would also be low in 34S. Under medium ƒO2
conditions, the δ34SΣs values of fluid producing the
pyrite-chalcopyrite-sphalerite assemblage are close to
those of pyrrhotite but smaller than those of pyrite, i.e.,
that δ34SΣs b δ34Spyrite (0.39‰), but approximately the
same as δ34Spyrrhotite (− 0.05‰).
The samples of ores and hydrothermal minerals can be
divided into three groups in accordance with their REE
distribution patterns. The REE distribution patterns of the
first group (Fig. 12B) may indicate that sedimentary rocks
in this region provided some ore-forming material. The
features of the second group (Fig. 12C) reflect involvement of lower crust or mantle material in gold mineralization. The features of the third group (Fig. 12D),
however, indicate that LREE/HREE fractionation occurred during evolution of ore-forming fluids, and that
HREE were significantly enriched during the late stage of
mineralization.
7.2.2. Lead isotopes
To characterize the tectonic environment of the source
region of ore lead, the Pb isotope data from various ore
deposits can be plotted against the different evolutionary
curves relative to Pb plumbotectonic models. As can
be seen from the 207Pb/204Pb–206 Pb/204 Pb diagram
(Fig. 17A), the data of hydrothermal sulfides and quartz
from the deposit mainly fall near the orogenic Pb
evolutionary curve, and between the orogenic and upper
crust Pb evolutionary curves, partly showing a transitional trend to the mantle source. In the 208Pb/204Pb–
206
Pb/204Pb diagram (Fig. 17B), the data for ore lead are
Samples were analyzed at the Stable Isotope Laboratory, Institute of
Mineral Resources, CAGS.
Sawaya'erdun deposit; results are listed in Table 8.
Because all quartz samples analyzed were products of
the main metallogenic stage in ore zone IV and the
analyses were carried out in the same laboratory as those
of Ye et al. (1999a), we report on the further treatment of
both sets of data in an attempt to acquire reliable
chronological data.
We conducted fitting treatment of all Rb–Sr isotopic
ratios in 19 quartz fluid inclusions using a least-squares
method. The slope of the acquired Rb–Sr isochron line
is equal to 0.004092 ± 0.000718, the Rb–Sr isochron
age, t, is 288 ± 50 Ma, and the initial (87Sr/86Sr)0 ratio is
0.714918 ± 0.000644. The Rb–Sr isochron of the 19
quartz fluid inclusions possesses good linearity (Fig. 16;
r = 0.9459), demonstrating that Sr isotopic homogenization occurred during metallogenesis.
7. Discussion
146
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Table 7
Ar/39Ar fast neutron activation analysis data of quartz from the Sawaya'erdun gold deposit
40
Incremental
heating
Heating
(40Ar/39Ar)m (36Ar/39Ar)m (37Ar/39Ar)m (38Ar/39Ar)m (39Ar)k10− 12mol
temperature (°C)
Conditions for sample 97sw–16: weight 0.40 g, irradiation parameter J 0.01219, irradiation time 55 h
6.63 × 1012 n/cm2.s, integrated flux of neutrons 1.32 × 1018 n/cm2
1
420
41.139
0.0728
0.1101
0.1835
0.37
2
550
27.674
0.0488
0.1142
0.1674
0.41
3
650
16.622
0.0216
0.1051
0.1270
0.86
4
730
13.947
0.0131
0.1310
0.1719
1.32
5
800
12.061
0.0069
0.1061
0.1473
3.04
6
880
12.921
0.0081
0.1575
0.2416
2.06
7
980
15.077
0.0161
0.2172
0.2692
1.51
8
1100
26.808
0.0468
0.5572
0.6596
0.54
9
1200
33.505
0.0593
0.6161
0.7371
0.45
10
1350
42.038
0.0764
0.6320
0.7325
0.36
11
1550
52.344
0.1016
0.5355
0.8437
0.21
39
Ark%
40
Ar/39Ark
( ± 1σ)
Appearance age
(Ma) ( ± 1σ)
47 mins., intermittent flux of neutrons
3.24
4.41
7.59
11.7
26.9
18.2
13.3
4.82
3.98
3.22
2.62
19.72 ± 1.01
13.21 ± 0.46
10.25 ± 0.16
10.07 ± 0.12
10.03 ± 0.09
10.28 ± 0.10
10.33 ± 0.14
13.10 ± 0.43
16.14 ± 0.67
19.63 ± 1.06
22.54 ± 1.64
388.61 ± 24.86
271.02 ± 8.10
212.36 ± 2.33
208.86 ± 1.63
208.07 ± 1.22
212.93 ± 1.44
214.00 ± 1.95
267.34 ± 7.55
324.09 ± 14.04
387.08 ± 25.90
438.04 ± 44.79
Conditions for sample 97sw–31: weight 0.02596 g, irradiation parameter J 0.010863, irradiation time 62 h, intermittent flux of neutrons 5 × 1012 n/cm2.s,
integrated flux of neutrons 1.08× 1018 n/cm2
1
380
63.888
0.16161
1.7172
0.61111
0.4588
5.35 16.50 ± 0.05 297.57 ± 14.26
2
460
41.016
0.10169
1.1837
0.46779
0.6838
7.98 11.20 ± 0.04 207.18 ± 7.27
3
540
27.573
0.05588
1.0946
0.26029
1.576
18.4
11.21 ± 0.02 207.39 ± 4.40
4
620
30.727
0.0663
0.63491
0.30363
1.275
14.8
11.24 ± 0.02 207.98 ± 4.98
5
700
32.261
0.0714
0.89707
0.35238
0.9737
11.3
11.31 ± 0.02 209.21 ± 5.65
6
820
35.833
0.0833
1.1997
0.43055
0.8345
9.74 11.41 ± 0.03 210.97 ± 6.78
7
950
44.5
0.10714
1.444
0.47142
0.6489
7.58 13.10 ± 0.04 240.16 ± 8.53
8
1100
45.312
0.10625
1.2348
0.47812
0.7417
8.66 14.16 ± 0.04 258.22 ± 9.26
9
1300
43.888
0.0972
1.1486
0.45277
0.8345
9.74 15.38 ± 0.03 278.89 ± 9.46
10
1500
63.695
0.15217
2.0776
0.71739
0.5328
6.22 19.13 ± 0.06 340.73 ± 18.23
Conditions for sample 97sw–32: weight 0.2669 g, irradiation parameter J 0.010863, irradiation time 62 h, intermittent flux of neutrons 5 × 1012 n/cm2.s,
integrated flux of neutrons 1.08× 1018 n/cm2
1
380
64.926
0.16097
1.9541
0.64878
0.4749
5.54 17.76 ± 0.05 318.26 ± 15.97
2
460
29.217
0.0608
0.95594
0.3113
1.333
15.5
11.38 ± 0.02 210.33 ± 5.11
3
540
29.09
0.0606
0.93164
0.30151
1.53
17.8
11.33 ± 0.02 209.43 ± 4.96
4
620
31.25
0.0687
1.3974
0.38749
1.112
12.9
11.14 ± 0.03 206.09 ± 6.00
5
700
35.694
0.0833
1.6303
0.53888
0.8342
9.74 11.32 ± 0.04 209.33 ± 8.19
6
820
41.438
0.10273
1.9781
0.64383
0.6765
7.9
11.39 ± 0.05 210.59 ± 9.79
7
950
47.148
0.11787
2.2316
0.6806
0.6092
7.11 12.68 ± 0.05 232.86 ± 11.47
8
1100
45.909
0.10606
2.0044
0.65151
0.7645
8.92 14.89 ± 0.05 270.64 ± 12.65
9
1300
57.959
0.14285
2.7759
0.83265
0.5673
6.62 16.20 ± 0.06 292.57 ± 17.54
10
1500
53.754
0.1228
2.0594
0.6807
0.6602
7.71 17.82 ± 0.05 319.39 ± 15.71
Samples were analyzed at the Isotope Laboratory, Institute of Geology and Geophysics, CAS.
all concentrated near the orogenic Pb evolutionary curve,
partly showing a transitional trend to the lower crust.
This demonstrates that ore lead is derived predominantly
from the crust, although the involvement of mantlesourced lead cannot be ruled out (Zheng et al., 2002).
7.2.3. Silica isotopes
Available studies have shown that variation of δ30Si
values is controlled by Si isotope dynamic fractionation.
The first precipitated silica has the lowest δ30Si value.
With increasing proportion of precipitated silica, δ30Si
values will increase steadily to a large, positive value
(Ding et al., 1994). Studies by Clayton (1986) also
indicate that when SiO2 is leached out of the rocks by
hydrothermal solutions, Si isotope fractionation will not
generally take place. However, when SiO2 is precipitated
from hydrothermal solutions, δ30Si fractionation between
precipitated SiO2 and the residual SiO2 in the hydrothermal solutions will occur, but to varying degrees. If
temperatures of hydrothermal solutions are relatively
high, SiO2 will be rapidly and completely precipitated,
and such Si isotope fractionation mentioned above would
be negligible. In contrast, if temperatures are relatively
low, SiO2 will be precipitated slowly and incrementally,
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Fig. 15. The 40Ar–39Ar age spectrum of Au-bearing quartz from the
Sawaya'erdun gold deposit. (A) 40Ar/39Ar age spectrum for sample
97sw-16 from ore zone no. IV. (B) 40Ar/39Ar age spectrum for sample
97sw-31 from ore zone no. IV. (C) 40Ar/39Ar age spectrum for sample
97sw-32 from ore zone no. IV.
and δ30Si values of SiO2 will be accordingly low. Formation temperatures of the Sawaya'erdun deposit are mainly
in the range 110∼220 °C, so the sedimentary (δ30Si =
−0.5 to +1.9‰) and volcanic rocks (δ30Si = +0.1‰) in
this region may have provided some silica for the oreforming fluids (Liu et al., 2002a; Zheng et al., 2002).
7.2.4. Carbon and oxygen isotopes
The δ13C values of CO2 in the fluid are highly variable. By comparing the variation ranges of δ13C values
of CO2 of various origins both in China and worldwide
(Dai, 1988; Dai and Chen, 1993), it is inferred that CO2
147
in the deposit may have two principal sources. One is
derived from dissolution of carbonate rocks and the
other is of endogenic inorganic origin. This feature is
more precisely reflected on the δ13C–δ18O plot (Fig. 18)
in which the C and O isotopic values of CO2 in crustal
fluids from three principal sources (organic matter,
marine carbonate rocks, and magma-mantle source) are
shown. The variation trends of C and O isotopic compositions of CO2 from the three sources, formed through
five major geological processes, are marked by arrows
(Liu and Liu, 1997; Mao et al., 2002; Liu et al., 2004b).
The plotted data (Fig. 18) indicate that CO2 might have
been derived not only from dissolution of marine
sedimentary carbonate rocks, but also from a magmatic
(mantle) source. In fact, the δ13C values of CO2 in fluid
inclusions are concentrated within two ranges: (1) − 10.5
to − 2.35‰ (mean − 5.75‰), probably reflecting involvement of a deep-source fluid; and (2) − 0.74 to
+ 4.98‰ (mean +1.63‰), approximating average δ13C
values of marine carbonate rocks (+ 0.56 ± 1.55‰; Keith
and Weber, 1964).
The δ13C values of CH4 in quartz fluid inclusions
(− 27.7 to − 23.9‰) are approximately the same as those
of CH 4 from hot springs distributed along the
Tengchong–Guohe fault zone in western Yunnan (ca.
− 20‰; Dai, 1988; Liu et al., 2001), the geothermal area
in Yellowstone, U.S.A. (− 28.4 to − 10.4‰; Macdonald
et al., 1983), hot water from Kamchatka, Russia (− 32.6
to − 21.4‰; Macdonald et al., 1983), and four
geothermal areas in Engava, New Zealand (− 29.5 to
− 24.4‰; Lyon and Hulston, 1984). They are also close
to the variation range of δ13C values (− 30 to − 20‰)
pointed out by Faure (1986), but differ from those of
CH4 from the closed interstices and caves of magmatic
rocks (− 14.6 to − 3.2‰; Zorikin et al., 1984) and from
free gases in magmatic rocks (− 19.3 to − 11.8‰;
Zorikin et al., 1984). The δ13C values of carbonaceous
material in the ore (− 25.5 to − 21.7‰; Zheng et al.,
1996) are also approximate to those of fluid. All this
indicates that CH4 in ore-forming fluid would probably
be derived chiefly from thermal decomposition of
organic matter. Theoretically, ΔCO2–CH4 should be
48.9∼26.3‰ (Friedman and O'Neil, 1977), bearing in
mind carbon isotope fractionation between CO2 and
CH4 in coexisting minerals at 100 to 280 °C, but
measured ΔCO2–CH4 (17.5∼27.4‰) is noticeably
smaller. This indicates a possibility that CH4 in the
ore-forming fluid was derived from CO2, or alternatively, from interaction between CO and H2 (degassing of
carbonate in sedimentary rocks and water/rock interactions between brines, carbonate and argillaceous rocks).
We, therefore, believe that the carbon in the deposit was
148
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Table 8
Results of Rb–Sr isotopic analyses for the Au-bearing quartz fluid inclusions from the Sawaya'erdun gold deposit
No.
Sample No.
Mineral
Rb/μg.g−1
Sr/μg.g−1
Rb/Sr
87
87
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
97sw12
97sw14
97sw16
97sw31
97sw32
97sw12-1
97sw31-1
97sw32-1
tc-01
tc-02
27kd-01
IV-Q1
IV-Q2
Q2
IV-I-02
IV-I-9
IV-I-11
IV-I-14
IV-I-07
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
Quartz
0.1159
0.3301
0.2145
0.2088
0.0308
0.119
0.2038
0.0309
0.514
0.341
0.14
0.441
1.825
0.468
0.483
1.037
0.258
0.232
0.262
3.297
2.291
3.372
3.339
4.007
3.287
3.394
3.978
2.405
2.072
1.822
2.131
1.533
1.963
1.478
4.467
2.498
1.626
2.408
0.035153
0.144086
0.063612
0.062534
0.007679
0.036203
0.060047
0.00776
0.213721
0.164575
0.076839
0.206945
1.190476
0.238411
0.326793
0.232147
0.103283
0.142681
0.108804
0.1015
0.4157
0.1835
0.1804
0.0222
0.1045
0.1732
0.0224
0.617
0.476
0.222
0.605
3.439
0.687
0.943
0.670
0.298
0.412
0.314
0.71582 ± 0.00005
0.71643 ± 0.00007
0.71543 ± 0.00005
0.71456 ± 0.00005
0.71374 ± 0.00006
0.71581 ± 0.00006
0.71457 ± 0.00001
0.7137 ± 0.00002
0.71759 ± 0.00001
0.71569 ± 0.00001
0.71734 ± 0.00001
0.71712 ± 0.00012
0.72885 ± 0.00002
0.71879 ± 0.00001
0.71771 ± 0.00003
0.71777 ± 0.00007
0.71793 ± 0.00001
0.71873 ± 0.00008
0.71631 ± 0.00017
Rb/86Sr
Sr/86Sr (2ó)
Note: Serial Nos. 1–8 samples were analyzed at the Isotope Analysis Center of the Yichang Institute of Geology and Mineral Resources, CAGS; data
for the others were published by Ye et al. (1999a).
partly derived from interactions between ore-forming
fluid and carbonate strata, and partly from the
decomposition of organic matter.
7.2.5. Hydrogen and oxygen isotopes
In order to explore the properties of ore-forming
fluids, our data (Table 5) were projected onto a diagram
of hydrogen and oxygen isotopic composition for various types of water (Fig. 19). It can be seen that the data
for quartz and siderite samples all fall outside the areas
of typical magmatic or metamorphic waters, indicating
that the ore-forming fluids consisted largely of groundwaters (meteoric waters). Some samples plot relatively
close to the meteoric water line; others lie at a greater
distance from the line. This so-called “δ18O shift” reflects a degree of O isotope exchange between meteoric
water and rock (Zhang, 1989; Liu et al., 1998b, 2000b;
Zheng et al., 2001, 2002).
various kinds. Such excess 40Ar includes low-temperature
cation-type and high-temperature anion-type excess 40Ar
(Li et al., 1995), or because excess argon is dominantly
hosted in fluid inclusions (because the fluid inclusions
contain K and Cl) and in mineral lattices (Wang and Qiu,
1999). The low-temperature excess argon contained in
fluid inclusions is usually released at 400 to 700 °C,
contributing a relatively small proportion of the total. But
the high-temperature excess 40Ar is chiefly hosted in
quartz and is usually released above 1000 °C. This portion
of excess 40Ar is relatively large (Sang et al., 1994;
Ying, 2002). The apparent ages obtained at high- and
low-temperature stages are both geochronologically
7.3. Age of mineralization
7.3.1. 40Ar/39Ar isotopic age
The saddle-shaped 40Ar/39Ar age spectrum indicates
that the analyzed samples contain excess 40Ar (Harrison
and McDougall, 1981; Zeitler and Gerald, 1986; Li et al.,
1995; Wang and Qiu, 1999; Ying, 2002; Ying and Liu,
2002). Generally, the low- and high-temperature apparent
ages of the saddle-shaped age spectrum are relatively
high, because of the release of obvious excess 40Ar of
Fig. 16. The Rb–Sr isochronal ages of quartz fluid inclusions from the
Sawaya'erdun gold deposits.
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
149
Fig. 17. Lead isotopic compositions of the Sawaya'erdun deposit (lead evolution model after Doe and Zartman, 1979; Zartman and Doe, 1981). M =
upper mantle lead; O = orogenic belt lead; U = upper crustal lead; L = lower crustal lead.
meaningless (Sang et al., 1994; Liu et al., 2002b, 2003a,
2004a). In contrast, ages of geological significance can
be acquired at the middle-temperature stage.
Since the 40Ar–39Ar age spectrum can be used to
discriminate the excess 40Ar in rock and minerals, the
plateau age or lowest apparent age in this kind of age
spectra is closer to the age of crystallization of the sample.
The true ages of some samples can be acquired (Lanphere
and Dalrymple, 1976; Zeitler and Gerald, 1986; Lanphere,
2000). Therefore, the lowest apparent age (206.09 ±
6.00∼208.07± 1.22 Ma) or the plateau age in the age
spectrum (208.33 ± 0.55∼210.59± 0.99 Ma) of Au-bearing quartz veins can be regarded as the upper age limit of
gold mineralization in the Sawaya'erdun gold deposits.
Included in the plateau ages are the age data acquired
at five heating stages from 460 to 980 °C; the amount of
released 39Ar accounts for 62.22∼77.69% of the total.
The data points involved in the plateau age calculation
can satisfy such pre-conditions that the sample was
deposited at the same time from the cognate source under
chemically closed condition (Folland, 1983; Zeitler and
Gerald, 1986; Sang et al., 1994; Li et al., 1995).
Therefore, isochron age calculations can be conducted.
Our results (Fig. 20) show that the plateau age (208.33 ±
0.55∼210.59 ± 0.99 Ma), the lowest apparent age
(206.09 ± 6.00∼208.07 ± 1.22 Ma) and the isochron age
(206.05 ± 2.38∼208.25 ± 3.39 Ma) are very close to each
other. This tells us that the determined ages of the quartz
samples may be both true and reliable and that the
plateau age represents the time of formation of quartz,
corresponding to Late Indosinian epoch. Furthermore,
initial (40Ar/39Ar)i ratios (294.9 ± 3.54∼310 ± 14.32 Ma)
are close to atmospheric value (295.5 ± 5), also indicating
that the analyzed samples contain no excess 40Ar or have
suffered an obvious 39Ar loss. This implies that excess
40
Ar have almost no impact on the analytical results and
that our plateau ages are considered both precise and
reliable. If the isochron intercept value significantly
150
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Fig. 18. δ18O versus δ13C diagram of siderite and fluid inclusions of quartz from the Sawaya'erdun deposits. For reference, the fields for typical
marine carbonates (MC), sedimentary organic matter (Sed Org), igneous carbonate (CMX, carbonatite and mantle xenoliths; BUR, basic and
ultrabasic rocks; G, granite) and mantle polyphase system (MM) are outlined. The arrows show typical isotopic trends resulting from carbonate
dissolution (Dis Carb), decarbonation (Dec), decarboxylation of organic matter (Decbx), oxidation of organic matter (Oxid Org), crystallization
differentiation (Crys Diff), influence of meteoric water (Influ MW), permeation of seawater (Perm SW), low tamperature alteration (LT alter) and
sedimentary hybridization trend (Sed Hybr). See text for further description. Modified after Liu and Liu (1997), Liu et al. (2004b).
differs from 295.5, then the analytical data should be
recalculated, in order to obtain the true age, using the
intercept value for the composition of non-radiogenic Ar
(Lanphere, 2000).
7.3.2. Rb–Sr isotopic ages on quartz fluid inclusions
We obtained a Rb–Sr isochron age of 288 ± 50 Ma on
quartz fluid inclusions from the deposit. However, it was
attested by the method of (87Sr/86Sr)0 − 1/Sr (Wendt and
Carl, 1991; Zheng, 1989) and isochron-fitted MSWD
(mean squared weighted deviation; Vollmer, 1976; Yang
et al., 2000) to be a mixed isochron age, and therefore
carries no chronological significance. Table 8 also shows
that 87Rb/86Sr is directly proportional to Rb/Sr and this
feature can be explained by the isochron model or the
mixing model (Qin, 1987). Considering the possibility
that there was mixing during the hydrothermal event, the
(87Sr/86Sr)0 − 1/Sr diagram was constructed (Fig. 21).
The data for samples constituting the isochron are mostly
not distributed around ISr = 0.714918 ± 0.000647, but
instead tend to show a linear distribution. This may
indicate that the samples constituting the isochron are,
for the most part, the result of mixing. Although the
isochron age obtained (T 288 ± 50 Ma, and ISr = 0.71491 ±
0.000647) may indeed be indicative of a mixed isochron,
we note with interest that this age is highly reminiscent of
that given for the Muruntau deposit.
Our studies have shown that the formation of the
Sawaya'erdun gold deposit is closely linked to mixing of
Fig. 19. δ18O versus δD diagram of fluid inclusions of the
Sawaya'erdun deposit (modified after Sheppard, 1986).
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
151
member is (87Rb/86Sr)0 = 0.711509 (Liu et al., 1999),
with Rb = 84.76 × 10 − 6 and Sr = 111.21 × 10 − 6 . This
value is the average value of 29 host rock samples
analyzed by NAA, including phyllite, siltstone and
siliceous rock samples (Liu et al., 2002a).
According to the slope of the mixing isochron, g =
[(( 87 Sr/ 86 Sr)A − ( 87 Sr/ 86 Sr)B) / k[(Rb/Sr)A − (Rb/Sr)B],
then k = 2.90 (for calculation procedure see Faure, 1986;
Qin, 1987; Zheng, 1989). The g value is figured to be
2.92925 × 10− 3 and the calculated age, 206 Ma. This
value represents closure time of the Rb–Sr isotopic
system or the time of material mixing.
7.4. Genesis of the deposit
Fig. 20. The 40Ar–39Ar isochronal ages of Au-bearing quartz from the
Sawaya'erdun deposits. (A) 40Ar/39Ar isochron diagram for sample
97sw-16. (B) 40Ar/39Ar isochron diagram for sample 97sw-31.
(C) 40Ar/39Ar isochron diagram for sample 97sw-32.
In the Sawaya'erdun region, both the Carboniferous
ore-hosting clastic rock formation and the underlying
Paleozoic strata are enriched in Au, Sb, and Hg (Ye
et al., 1998, 1999b; Zheng et al., 2001). This created one
favorable precondition for the formation of the deposit.
Since Paleozoic times, the southwestern Chinese
Tianshan region has experienced multiple episodes of
tectonism, resulting in complex tectonic deformation
and fragmentation of rocks (Zheng et al., 1996, 2001).
The locations, where the original stress energy was
large, have been converted to a depressurized dilatant
zone, providing an important locus for late hydrothermal activity and ore formation.
Late hydrothermal metallogenesis in the region is
closely associated with infiltration of meteoric water, so
understanding of the heat source is a key to understand
the forming mechanism of the deposit. It is commonly
accepted that gold metallogenesis in the southwestern
Tainshan region mainly occurred during the Late
Carboniferous–Permian (Goldfarb et al., 2001; Mao
et al., 2004). Since the Mesozoic, tectonomagmatic
infiltrating meteoric waters, which extracted ore-forming
elements from the ore-hosting and underlying strata,
with magmatic waters (Liu et al., 2002a). The diabase
dyke, a product of magmatic activity in the region, is
closely associated with gold metallogenesis, based on
our geological and geochemical observations (Liu et al.,
2002a). We therefore selected diabase dyke and host
rock in this region as the two end-members. The initial
condition of the diabase dyke end-member is (87Sr/
86
Sr)0 = 0.707815 (obtained by substituting the K–Ar
age, T = 207 Ma of this sample into the Rb–Sr isotopic
data), with Rb = 49.36 × 10− 6 and Sr = 150.80 × 10− 6,
which are the NAA average values of two diabase dyke
samples. The initial condition of the host rock end-
Fig. 21. (87Sr/86Sr)i versus 1/Sr diagram of quartz fluid inclusions from
the Sawaya'erdun deposit.
152
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
activity in this region was only weak (Wu et al., 1998;
Yang et al., 1999). Although there is no extensive
exposure of intrusive rocks in the studied region, some
diorite and diabase dykes have been found in the
Carboniferous and Permian strata (Zheng et al., 2001;
Liu et al., 2002a,b). This suggests that tectonomagmatic
activity was also intense in this region during the postPermian period. Available data show that Late Indosinian tectonomagmatic activity was not only intense but
also accompanied by formation of base- and precious
metal deposits in the southwestern Tianshan Mountains
(Liu et al., 2002b). Therefore, the tectonomagmatic
activities could not only provide some ore-forming
components, including Au and Sb for the Sawaya'erdun
gold deposit, but more importantly they provided a heat
source to drive the hydrothermal system, leading to
intensification of paleogeothermal activities during the
Late Indosinian. The mineralization age provides strong
evidence for formation of the Sawaya'erdun gold
deposit (and other gold deposits in southwestern
Chinese Tianshan Mountains) in the Late Indosinian.
This reflects that a certain indispensable key factor leads
to gold metallogeny and constrains deposit formation in
an integrated manner during the Late Indosinian in the
southwestern Chinese Tianshan Mountains.
Meteoric waters infiltrated downwards through
fractures, extracting ore-forming elements from Aubearing strata, forming mineralizing hydrothermal solutions. Density tends to decrease with rise in temperature.
When these solutions reached a certain critical depth,
they would return upwards into a favorable depressurized space, whereas the change of physico-chemical
conditions (temperature, pressure, acidity-alkalinity and
redox potential) would increase considerable concentrations of ore-forming materials, such as gold and base
metals in the residual solutions after the crystallization of
quartz. Sulfides and native gold would crystallize from
the saturated solutions to form the economic orebodies.
8. Concluding remarks and comparison with
Muruntau
The Sawaya'erdun gold deposit is the first case of a
large, low-grade orogenic gold deposit in the Chinese
portion of the Tianshan Mountains. In some ways, it
can be compared with the Muruntau gold deposit in
Uzbekistan with respect to geological setting, host
lithology, mineralization style, mineral assemblages,
geochemical association, metallogenic processes involved — and possibly also age.
The common features shared by the Sawaya'erdun
and Muruntau are as follows.
(1) The regional geotectonic setting of the Sawaya'erdun deposit is consistent with that of the Muruntau
gold deposit in adjacent Uzbekistan (Kotov and
Poritskaya, 1993; Drew et al., 1996). Both deposits
are controlled by the Turkestan suture zone (Cai
et al., 1993; Mao et al., 2004).
(2) Lithological characteristics of the ore-hosting strata
are a suite of typical turbidities (Zheng et al., 1996,
2001) in both cases, although the age of the orehosting strata is Late Carboniferous for Sawaya'erdun (Liu et al., 1999) and is either Ordovician–
Silurian (according to Mukhin et al., 1988; Uspenskiy and Aleshin, 1993; Loshchinin, 1996; Kempe
et al., 2001), Cambrian–Ordovician (according to
Mukhin et al., 1988; Cai et al., 1993; Tan, 1993), or
Precambrian–Lower Paleozoic (according to
Drew et al., 1996; Skryabin et al., 1997) for
Muruntau.
(3) Mineralization is controlled by fracture zones, and
orebodies are composed of ore-bearing quartz
veins or quartz–siderite veinlets or network veins.
The ore minerals and the ore types in the veins are
comparable.
(4) There exist a large number of base metal sulfides
and bismuth minerals associated with gold minerals in both deposits. Mineral assemblages differ
from those of Carlin-type gold deposits, which are
characterized by a low-temperature mineral assemblage of arsenopyrite-(As) pyrite-realgar-orpiment-cinnabar (Arehart, 1996; Liu et al., 2000d,
e, 2003b).
(5) The extensive Au–As–Sb–Hg–Ag–Bi geochemical anomalies over a very large area in the
Sawaya'erdun field, with perfect element association and distinct zonation (Liu et al., 1998a) are
also consistent with those in the Muruntau deposit
(Skryabin et al., 1997).
(6) The Muruntau deposit is characterized by the
presence of visible gold (generally within the
range 0.001 to 1 mm), which is present as microfine enclaves scattered in quartz and sulfides. The
gold also occurs as micro-fine veinlets (Cai et al.,
1993; Tan, 1993). Abundant microscopic gold and
gold veinlets were also recognized in the
Sawaya'erdun deposit (Zheng et al., 1996; Liu
et al., 2000a; Zheng et al., 2001). Recovered gold
is distributed in the fragmented parts of quartz
and arsenopyrite or pyrite or at the rims of the
mineral grains.
(7) The Sawaya'erdun deposit is generally similar to
the Muruntau deposit with respect to its multiepisode and multi-stage metallogenesis.
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
The Sawaya'erdun gold deposit also possesses some
unique features.
(1) The ore-forming temperature is low. As stated
above, with the exception of the ore-barren quartz
stage at which the temperature was relatively high
(220 to 320 °C), homogenization temperatures of
quartz fluid inclusions associated with metallogenesis are all below 223 °C, mostly within the
range 120 to 210 °C. However, with the exception
of the fourth metallogenic stage at Muruntau,
during which the temperature was relatively low
(Zairi and Kurbanov, 1992), the average temperatures for the other three metallogenic stages all
exceed 220 °C (generally in the range of 220 to
350 °C, even up to as much as 500 °C), significantly higher than any temperature estimate
for Sawaya'erdun.
(2) The geochemical signatures are somewhat different. Although the Sawaya'erdun and Muruntau
gold deposits are both characterized by high Ag
anomalies, the silver mineralogy appears to be
substantially different. Selenides and tellurides
have been widely found in Muruntau (e.g., Drew
et al., 1996), but in the Sawaya'erdun gold deposit
no selenide or telluride mineral has yet been found.
In addition, no high W anomalies have been found
at Sawaya'erdun, yet Sb is much more abundant
with Sb lodes defined. As for gold minerals,
Muruntau deposit is dominated by native gold of
high fineness (890 to 910; Cai et al., 1993; Liu
et al., 1998a), whereas the Sawaya'erdun deposit is
dominated by electrum with fineness of only 700
to 800 (Zheng et al., 1996; Liu et al., 2000a; Zheng
et al., 2001). Formation of electrum is controlled
by a number of dynamic factors such as element
migration style, the extent of opening or closeness
of the system and the precipitation mechanism of
electrum itself (Wilkinson et al., 1999). Generally
speaking, electrum is formed under low temperature conditions (e.g., Shelton et al., 1990), but can
also be formed under high temperature conditions
(Hagemann et al., 1998).
There is still uncertainty about whether formation of
the Sawaya'erdun and Muruntau gold deposits was
synchronous. Previous K–Ar and Rb–Sr dating of
hydrothermal alteration and sulfide–quartz veins yielded
four-phase ages from 272.6, 256.2, 230.2, and 219.4 Ma
for the Muruntau gold deposit (Kotov and Poritskaya,
1992; Marakushev and Khokhlov, 1992). Kostitsyn
(1993, 1996) obtained also Rb–Sr isochrons of density
153
separates from auriferous quartz vein samples (include
quartz-tourmaline vein, quartz-arsenopyrite veinlet and
Ag–Au quartz-adularia veins) in the Muruntau of 257 ±
13, 230 ± 3.5, and 219 ± 4.2 Ma. Recently, Wilde et al.
(2001) obtained 39Ar–40Ar analytical results of stockpiled ore samples in the Muruntau deposit, which
provided evidence of two sericite-forming events at
about 245 and 220 Ma. Kempe et al. (2001) obtained
Sm–Nd isochron of scheelite from steeply dipping veins
in the Muruntau of 279 ± 18 Ma, which is interpreted to
date high-grade ore formation in the deposit. The span of
ages for Muruntau is thus from 219 to 279 Ma (Kotov
and Poritskaya, 1992; Marakushev and Khokhlov, 1992;
Kostitsyn, 1993, 1996; Kempe et al., 2001; Wilde et al.,
2001). The formation age of the Sawaya'erdun deposit
is younger — if we exclude the Rb–Sr isochron age
of 288 ± 50 Ma discussed in Section 7.3.2 above.
39
Ar–40Ar plateau and isochron ages of auriferous
quartz samples range from 206 to 213 Ma, and Rb–Sr
ages of quartz fluid inclusions give 206 Ma. Such ages
are consistent with the ages for the Dashankou gold
deposit (207.16 ± 0.85∼212.59 ± 0.68 Ma; Liu et al.,
2004a), and the Bulong gold deposit (212.18 ± 0.83 Ma;
Liu et al., 2004b), but is nevertheless consistent with the
age of the fourth-phase age given for Muruntau (219.4 ±
4.2 Ma; Kotov and Poritskaya, 1992; Marakushev and
Khokhlov, 1992). It also agrees closely with the third of
the three Rb–Sr ages (219 ± 4.2 Ma) given by Kostitsyn
(1993, 1996) or one of the two 39Ar–40Ar ages (220 Ma)
of Wilde et al. (2001) from Muruntau. Both isotopic
systems apparently record a succession of hydrothermal
events.
Acknowledgements
Funding for this project was jointly granted by the
111 project (Grant No. B07011), by the National Natural
Science Foundation of China (Grant No. 40073019,
40234051, and 40434011) and by the State No. 305
Program of Solving Key Problems in Science and
Technology (No.95-04-03-01). Thanks are due to
Chief Engineers Zhang Liangcheng, Liu Dequan and
Wang Futong from the Xinjiang Bureau of Geology
and Mineral Resources for their contribution to
fieldwork and to Senior Engineers Luo Zhiling and
Wang Jin from the No.2 Geological Party of Xinjiang
Bureau of Geology and Mineral Resources for their
great help. This manuscript has also benefited from
the comments and critical reviews from Ore Geology
Reviews referee Andy Wilde and Associate Editor
Alexander Yakubchuk, who has also handled the
manuscript.
154
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
References
Arehart, G.B., 1996. Characteristics and origin of sediment-hosted disseminated gold deposit: a review. Ore Geology Reviews 11, 383–403.
Cai, H., Zheng, Y., Li, F., 1993. Geological investigation of Muruntau
gold deposit. Mineralogy and Geology 7, 408–413 (in Chinese
with English abstract).
Chen, H., 1994. Ore Isotopic Geochemistry. Zhejiang University
Press, Hangzhou, China. 340 pp. (in Chinese).
Chen, F., Li, H., 2003. Metallogenic Chronology of the Sawayaerdun
Gold–Antimony Deposit in Xinjiang. Acta Geoscientia Sinica 24,
563–567 (in Chinese with English abstract).
Clayton, R.N., 1986. High temperature isotope effects in the early solar
system. Stable isotopes in high temperature geological processes.
In: Valley, J.W., Taylor Jr., O'Neil, J.R. (Eds.), Reviews in
Mineralogy, vol. 16, pp. 129–139.
Clayton, R.N., Mayeda, T.K., 1963. The use of bromine pentafluoride
in the extraction of oxygen from oxides and silicates for isotopic
analysis. Geochimica et Cosmochimica Acta 27, 43–52.
Cliff, R.A., Janes, G., Choi, W.C., Lee, T.J., 1985. Strontium isotopic
equilibration during metamorphism of tillites from Ogcheon belt,
South Korea. Contributions to Mineralogy and Petrology 90,
346–352.
Coleman, M.L., Sheppard, T.J., Durham, J.J., Rouse, J.E., Moore, G.R.,
1982. Reduction of water with zinc for hydrogen isotope analysis.
Analytical Chemistry 54, 993–995.
Dai, J., 1988. Carbon isotopic composition characteristics and origin
of natural gases at Liuhuangtang, Tengchong County, Yunnan
Province. Chinese Science Bulletin 33, 1168–1170 (in Chinese).
Dai, J., Chen, Y., 1993. Carbon isotopic characteristics of alkane
fraction in biogenic gases in China and their discriminating
indicators. Science in China. Series B, Chemistry, Life Sciences &
Earth Sciences 23, 303–305 (in Chinese).
Ding, T., Jiang, S., Wan, D., Li, Y., Li, J., Song, H., Liu, Z., Yao, X.,
1994. Silicon Isotope Geochemistry. Geology Press, Beijing.
102 pp. (in Chinese).
Dodson, M.H., 1976. Kinetic processes and thermal history of slowly
cooling solids. Nature 259, 551.
Doe, B.R., Zartman, R.E., 1979. Plumbotectonics. In: Barnes, H. (Ed.),
Geochemistry of Hydrothermal Ore Deposits. John Wiley and
Sons, New York, pp. 22–66.
Drew, L.J., Berger, B.R., Kurbanov, N.K., 1996. Geology and
structural evolution of the Muruntau gold deposit, Kyzylkum
Desert, Uzbekistan. Ore Geology Reviews 11, 175–196.
Fang, W., Hu, R., Xie, G., Su, W., Qi, L., 2001. Diageneticmetallogenic ages of pyritic cherts and their implications in
Mojiang nickel-gold deposit in Yunnan Province, China. Chinese
Science Bulletin 46, 1823–1827.
Faure, G., 1986. Principles of Isotope Geology, 2nd ed. John Wiley
and Sons, New York. 589 pp.
Folland, K.A., 1983. 40Ar/39Ar incremental heating plateaus for biotite
with excess argon. Isotope Geoscience 1, 3–21.
Friedman, I., O'Neil, J.R., 1977. Compilation of stable isotope
fractionation factors of geochemical interest. Data Geochemistry,
In: Fleischer, M. (Ed.), 6th ed. United States Geological Survey
Professional Paper, Washington DC, vol. 440-KK. 111 pp.
Goldfarb, R.J., Groves, D.I., Gardoll, S., 2001. Orogenic gold and
geologic time: a global synthesis. Ore Geology Reviews 18, 1–75.
Hagemann, S.G., Brown, P.E., Ridley, J., Stem, P., Fournelie, J., 1998.
Ore petrology, chemistry and timing of electrum in the Archean
Hypozonal lode gold deposit, Western Australia. Economic
Geology 93, 271–291.
Harrison, T.M., McDougall, I., 1981. Excess 40Ar in metamorphic
rocks from Broken Hill, New South Wales: implications for
40
Ar/39Ar age spectra and the thermal history of the region. Earth
and Planetary Science Letters 55, 123–149.
Hollister, L.S., 1982. Metamorphic evidence for rapid (2 mm/yr) uplift
of a portion of the central gneiss complex, Coast Mountains.
Canadian Mineralogist 20, 319–332.
Humphrises, F.J., Cliff, R.A., 1982. Sm–Nd dating and cooling history
of Scourian granulites, Sutherland. Nature 295, 515.
Jiang, B., 1989. Methodology and its limitation of Archean rock dating.
Geochemica 18, 103–120 (in Chinese with English abstract).
Keith, M.L., Weber, J.N., 1964. Carbon and oxygen isotopic
composition of selected limestone and fossils. Geochimica et
Cosmochimica Acta 28, 1787–1816.
Kempe, U., Belyatsky, B.V., Krymsky, R.S., Kremenetsky, A.A.,
Ivanov, P.A., 2001. Sm–Nd and Sr isotope systematics of scheelite
from the giant Au(–W) deposit Muruntau (Uzbekistan); implications for the age and sources of Au mineralization. Mineralium
Deposita 36, 379–392.
Knauth, I.P., Epstein, S., 1976. Hydrogen and oxygen isotope ratios in
nodular and bedded cherts. Geochimica et Cosmochimica Acta 40,
1095–1110.
Kostitsyn, Y.A., 1993. Rb–Sr isotope study of the Muruntau deposit:
dating of ore veins by the isochron technique. Geokhimiya 31,
1308–1319 (in Russian).
Kostitsyn, Y.A., 1996. Rb–Sr isotope study on the Muruntau deposit:
magmatism, metamorphism, and mineralisation. Geokhimiya 34,
1123–1138 (in Russian).
Kotov, N.V., Poritskaya, L.G., 1992. The Muruntau gold deposit: its
geologic structure, metasomatic mineral associations and origin.
International Geology Review 34, 77–87.
Kotov, N.V., Poritskaya, L.G., 1993. The Muruntau gold deposit.
Mineralogy in the service of mankind, Programme and Abstracts.
International Congress on Applied Mineralogy, 1993 Annual
Meeting, Freemantle, Australia, pp. 190–193.
Lanphere, M.A., 2000. Comparison of conventional K–Ar and
40
Ar/39Ar dating of young mafic volcanic rocks. Quaternary
Research 53, 294–301.
Lanphere, M.A., Dalrymple, G.B., 1976. Identification of excess 40Ar
by the 40Ar/39Ar age spectrum technique. Earth and Planetary
Science Letters 32, 141–148.
Laznicka, P., 1983. Giant ore deposit: a quantitative approach. Global
Tectonics and Metallogeny 2, 41–63.
Laznicka, P., 1989. Derivation of giant ore deposits. 28th International
Geological Congress, Washington DC, United States, vol. 2,
pp. 2268–2269. Abstracts.
Li, X., 1996. A discussion on the model and isochron ages of Sm–Nd
isotopic systematics: suitability and limitation. Scientia Geologica
Sinica 31, 97–104 (in Chinese with English abstract).
Li, X., Luo, W., 1997. Geological character of Sawaya'erdun gold deposit
in Xinjiang: the first one of Muruntau-type gold deposit in China.
Acta Geologica Gansu 6, 62–66 (in Chinese with English abstract).
Li, H., Liu, J., Wei, L., 1993. Fluid Inclusion Chronology of
Hydrothermal Ore Deposits and its Geological Reply. Geological
Press, Beijing, pp. 22–41.
Li, Z., Dai, T., Qiu, H., 1995. The release of 40Ar(40Ar⁎ + 40ArE) and 39Ar
in 40Ar–39Ar samples and its chronological significance. Chinese
Journal of Geology 30, 40–45 (in Chinese with English abstract).
Liu, J., Liu, J., 1997. Basinal fluid genetic model of fine disseminated
gold deposits in the Golden Triangle area between Yunnan,
Guizhou and Guangxi. Acta Mineralogica Sinica 17, 448–456 (in
Chinese with English abstract).
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Liu, D., Tang, Y., Zhou, N., 1998a. The direction of search for minerals
of the Muruntau-type gold deposits in Xinjiang. Gold Science &
Technology 6, 18–22 (in Chinese with English abstract).
Liu, J., Zheng, M., Zhou, D., Liu, J., Zhou, Y., Gu, X., Zhang, B., Lin,
L., 1998b. The components and evolution of the hydrogen and
oxygen isotopes of ore-forming fluids from Laerma gold ore belt.
Journal of Changchun University of Science and Technology 28,
93–98 (in Chinese with English abstract).
Liu, J., Zheng, M., Long, X., Gu, T., 1999. Reidentification of the orehosted strata age of the Sawaya'erdun gold deposit in Xinjiang,
southwestern Tianshan. Chinese Science Bulletin 44, 1811–1814.
Liu, J., Zheng, M., Long, X., Zhang, S., Yin, H., Wang, J., 2000a.
Identification of Sawaya'erdun Muruntau gold deposit in Xinjiang,
southwestern Tianshan Mountains Mountains and its signification.
Journal of Precious Metallic Geology 9, 129–132 (in Chinese with
English abstract).
Liu, J., Li, C., Pan, J., Hu, R., Liu, X., Zhang, Q., 2000b. Isotopic
geochemistry in the copper deposits from sandstone and shale in
Lanping-Simao basin, western Yunnan. Mineral Deposits 19,
223–234 (in Chinese with English abstract).
Liu, J., Zheng, M., Liu, J., Zhou, D., 2000c. Sulfur isotopic
composition and its geological significance of the Cambrian gold
deposits in western Qinling, China. Journal of Changchun
University of Science and Technology 30, 150–156 (in Chinese
with English abstract).
Liu, J., Liu, J., Zheng, M., Liu, X., 2000d. The Au–Se paragenesis in
Cambrian stratabound gold deposits, Western Qinling Mountains,
China. International Geology Review 42, 1037–1045.
Liu, J., Zheng, M., Liu, J., Su, W., 2000e. Geochemistry of the Laerma
and Qiongmo Au–Se deposits in the western Qinling Mountains,
China. Ore Geology Reviews 17, 91–111.
Liu, J., Li, C., Zhang, Q., Pan, J., Liu, Y., Liu, X., Liu, S., Yang, W.,
2001. Wood textures in Jinman copper deposits in western Yunnan.
Science in China, Series D: Earth Sciences 44, 545–554.
Liu, J., Zheng, M., Long, X., Yin, H., Wang, J., Li, E., 2002a.
Metallogenic characteristics of Sawaya'erdun gold deposit in
Xinjiang. Acta Mineralogica Sinica 22, 53–62 (in Chinese with
English abstract).
Liu, J., Zheng, M., Long, X., Li, E., Wang, J., Sang, H., Yin, H.,
2002b. The metallogenetic age of Sawaya'erdun gold deposit in
Xinjiang, southwestern Tianshan Mountains. Journal of Mineralogy and Petrology 22, 19–23 (in Chinese with English abstract).
Liu, J., Li, Z., Zhang, Q., Liu, Y., Li, C., He, M., Sang, H., 2003a. The
40
Ar–39Ar fast-neutron activation age of quartz from Jinman
copper deposit, western Yunnan. Chinese Journal of Geology 38,
529–531 (in Chinese with English abstract).
Liu, J., Zheng, M., Liu, J., Gu, X., Zhou, Y., Feng, C., 2003b.
Mechanical transport of metallogenic materials in hydrothermal
solutions: Evidence from the microspherules in micro-disseminated gold deposits, northwestern Sichuan, China. Ore Geology
Reviews 22, 1–16.
Liu, J., Li, E., Long, X., Zheng, M., Wang, J., Sang, H., 2004a. The
metallogentic age of Dashankou gold deposit in Xinjiang, southwestern Tianshan Mountains. Journal of Jilin University (Earth
Science Edition) 34 (1), 137–143 (in Chinese with English abstract).
Liu, J., He, M., Li, Z., Liu, Y., Li, C., 2004b. The oxygen and carbon
isotopic geochemistry and its significance in Baiyangping silvercopper polymetallic metallogenic concentration area, Lanping Basin.
Mineral Deposits 23, 1–10 (in Chinese with English abstract).
Long, X., Zheng, M., Zhang, S., Liu, J., 1998. Study on fluid
inclusions of Sawaya'erdun gold deposit. Mineral Deposits 17,
973–976 (suppl., in Chinese).
155
Loshchinin, V.P., 1996. The problem of gold genesis at the
Ordovician–Silurian deposits of Muruntau ore field (Western
Uzbekistan). 30th International Geological Congress, Beijing,
China, vol. 1/3, p. 780. Abstract.
Ludwig, K.R., 1996. ISOPLOT: a plotting and regression program for
radiogenic-isotope data (Version 2.90). United States Geological
Survey Open-File Report 91 (445), 1–47.
Lyon, G.L., Hulston, J.R., 1984. Carbon and hydrogen isotopic
compositions of New Zealand geothermal gases. Geochimica et
Cosmochimica Acta 48, 1161–1171.
Macdonald, G.A., Abbott, A.T., Peterson, F.L., 1983. Volcanoes in the
Sea, the Geology of Hawaii, 2nd ed. University of Hawaii Press,
Honolulu. 517 pp.
Mao, J., He, Y., Ding, T., 2002. Mantle fluids involved in
metallogenesis of Jioadong (east Shandong) gold district: evidence
of C, O and H isotopes. Mineral Deposits 21, 121–127 (in Chinese
with English abstract).
Mao, J., Konopelko, D., Seltmann, R., Lehmann, B., Chen, W., Wang,
Y., Eklund, O., Usubaliev, T., 2004. Postcollisional age of the
Kumtor gold deposit and timing of Hercynian events in the Tien
Shan, Kyrgyzstan. Economic Geology 99, 1771–1780.
Marakushev, A.A., Khokhlov, V.A., 1992. A petrological model for
the genesis of the Muruntau gold deposit. International Geology
Review 34, 59–76.
McCrea, J.M., 1950. The isotopic chemistry of carbonates and
a paleotemperature scale. Journal of Chemical Physics 18,
849–857.
Mukhin, P.A., Savchuk, Y.S., Kolesnikov, A.V., 1988. The position of
the “Muruntau lens” within the metamorphic rock unit of the
Southern Tamdytau (Central Kyzylkum Desert). Geokhimiya 2,
64–72 (in Russian).
Ohmoto, H., 1972. Systematics of sulfur and carbon isotopes in
hydrothermal ore deposits. Economic Geology 67, 551–578.
O'Neil, J.R., Silberman, M.L., 1974. Stable isotope relations in
epithermal Au–Ag deposits. Economic Geology 69, 902–909.
Qin, Z., 1987. The mixing isochron model and its significance for
isotopic chronology. Science in China. Series B, Chemistry, Life
Sciences & Earth Sciences 30, 95–103.
Robin, M.B., Joaquin, R., Spencer, R.T., Richard, M.T., Joseph, L.W.,
1999. Lead isotope compositions of Late Cretaceous and Early
Tertiary igneous rocks and sulfide minerals in Arizona: implications for the sources of plutons and metals in porphyry copper
deposits. Economic Geology 94, 211–244.
Rui, Z.Y., Goldfarb, R.J., Qiu, Y.M., Zhou, T.H., Chen, R.Y., Pirajno,
F., Yun, G., 2002. Paleozoic — early Mesozoic gold deposits of the
Xinjiang Autonomous Region, northwestern China. Mineralium
Deposita 37, 393–418.
Sang, H., Wang, S., Hu, S., Qiu, Ji., 1994. 40Ar/39Ar dating method
and Ar isotopic mass spectrometry analysis of quartz. Journal of
Chinese Mass Spectrometry Society 15, 17–27 (in Chinese with
English abstract).
Seltmann, R., Shatov, V., Yakubchuk A., eds., 2003. Mineral deposits
database and thematic maps of Central Asia (scale 1:1,500,000),
Explanatory Notes to GIS ArcView 3.2 and MapInfo v7 package:
London, Centre for Russian and Central Asian Mineral Studies,
Natural History Museum, 116 pp.
Shayakobov, T., Kremenetsky, A., Mintser, E., Obraztsov, A.,
Graupner, T., 1999. The Muruntau orefield. In: Islamov, F.,
Kremenetsky, A., Seltmann, R. (Eds.), Au, Ag, and Cu deposits of
Uzbekistan. Excursion Guidebook of the IGCP-473 International
Field Conference to Uzbekistan, 27 Aug. – 4 Sept. 1999. IAGOD
Guidebook Series, vol. 7. GFZ, Potsdam, pp. 37–74.
156
J. Liu et al. / Ore Geology Reviews 32 (2007) 125–156
Shelton, K.L., So, C.S., Haeussler, G.T., Chi, S.J., Lee, K.Y., 1990.
Geochemical studies of the Tongyoung gold–silver deposits,
Republic of Korea. Evidence of meteoric water dominance in a Tebearing epithermal system. Economic Geology 85, 1114–1132.
Sheppard, S.M.F., 1986. Characterization and isotopic variations in
natural waters. In: Valley, J.W., Taylor Jr., H.P., O'Neil, J.R. (Eds.),
Stable Isotopes in High Temperature Geological Processes.
Reviews in Mineralogy, vol. 16, pp. 165–183.
Skryabin, V.F., Guryanov, V.F., Dzhurayev, A.D., 1997. Primary geochemical zoning at Muruntau ore deposit. Razvedka i Okhrana Nedr
(Exploration and Conservation of Resources) 11, 11–16 (in Russian).
Tan, G., 1983. A Concise Course in Paleontology. Geological Press,
Beijing. 306 pp. (in Chinese).
Tan, K.R., 1993. Proceeding of superdeep drilling geology in
Muruntau gold field. Gold Science & Technology 1, 17–28 (in
Chinese with English abstract).
Tu, G., 2000. The Super-Large Ore Deposits in China. Science Press,
Beijing, China. 266 pp.
Uspenskiy, Y.I., Aleshin, A.P., 1993. Patterns of scheelite mineralization in the Muruntau gold deposit, Uzbekistan. International
Geology Review 35, 1037–1051.
Vollmer, R., 1976. Rb–Sr and U–Th–Pb systematics of alkaline rocks:
the alkaline rocks from Italy. Geochimica et Cosmochimica Acta
40, 283–295.
Wang, S., Qiu, J., 1999. A study of saddle-shaped 40Ar–39Ar age
spectrum of chert as exemplified by the Tieling Formation, Jixian
section. Acta Geoscientia Sinica 20, 363–367 (in Chinese with
English abstract).
Wang, Y., Yang, J., Tao, X., 1988. Experimental research on Sm–Nd
isotopic method of fossil-mineral-rock and its application. Journal
of Nanjing University (Science Edition) 20, 297–308 (in Chinese
with English abstract).
Wang, C.Y., Zhou, M.K., Yan, Y.J., Wu, Y.L., Zhao, Y.G., Qian, Y.Z.,
2000. Lower Devonian conodonts from the Sawaya'erdun gold
mine area, Wuqia (Ulugqat) county, Xinjiang. Acta Micropalaeontologica Sinica 17, 255–264.
Wendt, I., Carl, C., 1991. The statistical distribution of the mean
squared weighted deviation. Chemical Geology 86, 275–285.
Wilde, A.R., Layer, P., Mernagh, T., Foster, J., 2001. The giant Muruntau
gold deposit: geologic, geochronlogic, and fluid inclusion constraints
on ore genesis. Economic Geology 96, 633–644.
Wilkinson, J.J., Boyce, A.J., Earls, G., Fallick, A.E., 1999. Gold
remobilization by low-temperature brines: evidence from the
Curraghinalt gold deposit, Northern Ireland. Economic Geology
94, 289–296.
Wu, R., Tian, C., Huang, M., Cai, H., Zhi, Q., Qi, S., Cao, L., Wang,
R., 1998. Geological characteristics of the gold–copper deposits in
the western Tianshan. Journal of Precious Metallic Geology 7,
1–19 (in Chinese with English abstract).
Yakubchuk, A., Cole, A., Seltmann, R., Shatov, V., 2002. Tectonic
setting, characteristics, and regional exploration criteria for gold
mineralization in the Altaid orogenic collage: the Tien Shan
province as a key example. Society of Economic Geologists
Special Publication 9, 177–201.
Yang, F., Ye, Q., Fu, X., Ye, J., 1999. Distribution and metallogenic
conditions of gold deposit in the southernwest Tianshan Mountains. Xinjiang Geology 17, 129–136 (in Chinese with English
abstract).
Yang, J., Zhou, X., Chang, H., 2000. The Rb–Sr isochrology and
metallogenic ages of Au-bearing minerals and ores in the Linlong
gold deposit, Shandong. Chinese Science Bulletin 45, 1547–1553
(in Chinese).
Ye, J., Ye, Q., Meng, X., Wang, J., Zhang, X., 1998. The forming
condition and prospecting of Muruntau type gold deposit in
southwestern Tianshan, China. Mineral Deposits 17, 41–42
(suppl., in Chinese).
Ye, J., Wang, L., Ye, Q., Li, H., Zhang, X., 1999a. Metallogenic epoch
and ore-bearing strata age of the Sawaya'erdun gold–antimony
deposit in Tianshan Mountains, China. Acta Geoscientia Sinica 20,
278–283 (in Chinese with English abstract).
Ye, J., Ye, Q., Wang, J., Wu, Y., Yang, F., Fu, X., 1999b. Geochemistry
and metallogenic mechanism of the Sawaya'erdun gold–antimony
deposit. Mineral Deposits 18, 64–72 (in Chinese with English
abstract).
Ye, Q., Ye, J., 1998. The mechanism and origin of Sawaya'erdun gold
and antimony deposit in Xinjiang, China. Mineral Deposits 17,
287–290 (suppl., in Chinese).
Ying, H., 2002. 40Ar–39Ar fast neutron activation dating of gold-rich
quartz vein from the Mojiang nickel-gold deposit, Yunnan.
Chinese Journal of Geology 37, 107–109 (in Chinese with English
abstract).
Ying, H., Liu, B., 2002. 40Ar–39Ar dating of gold-bearing quartz veins
in the Fengyang and Zhangbaling areas, Anhui Province, and its
geological significance. Mineral Deposits 21, 240–245 (in Chinese
with English abstract).
York, D., 1969. Least squares fitting of a straight line with correlated
errors. Earth and Planetary Science Letters 5, 320–324.
Yu, C.C., Lin, Y.D., Shi, Y., Huang, Z.X., Yu, X.G., 1983.
Carboniferous and Permian Corals. Jilin People's Publishing
House, Changchun. 357 pp. (in Chinese).
Zairi, N.M., Kurbanov, N.K., 1992. Isotopic-geochemical model of ore
genesis in the Muruntau ore field. International Geology Review
34, 88–94.
Zartman, R.E., Doe, B.R., 1981. Plumbotectonics — the model.
Tectonophysics 75, 135–162.
Zeitler, P.K., Gerald, D.F., 1986. Saddle-shaped 40Ar/39Ar age spectra
from young, micro-structurally complex potassium feldspars.
Geochimica et Cosmochimica Acta 50, 1185–1199.
Zhai, Y.S., 2006. Innovative research on mineral deposit geology.
Earth Science Frontiers 13, 1–7.
Zhang, L., 1989. Petrogenetic and Mineragenetic Theories and
Prospecting. Beijing University of Industry Press, Beijing. 200 pp.
(in Chinese).
Zheng, Y., 1989. Influences of the nature of the initial Rb–Sr system
on isochron. Chemical Geology. Isotope Geoscience Section 80,
1–16.
Zheng, M., Liu, J., Long, X., Tian, Z., 1996. Metallogenical
Conditions and Prospects of Muruntau-Type Gold Deposits in
the Southern Tianshan, Xinjiang. Chengdu University of Technology. 113 pp. (in Chinese).
Zheng, M., Liu, J., Long, X., Zhang, S., 1998. The first case of
Muruntau-type gold deposit in China. Mineral Deposits 17,
381–384 (suppl., in Chinese with English abstract).
Zheng, M., Zhang, S., Liu, J., Long, X., Song, X., 2001. The Forming
Setting of Muruntau-Type Gold Deposit in Southwestern Tianshan
Mountains and Metallogenic Exploration. Geology Press, Beijing.
131 pp. (in Chinese).
Zheng, M., Liu, J., Zhang, S., Long, X., 2002. Isotopic composition
and genetic indication of Sawaya'erdun gold deposit, Xinjiang.
Journal of Chengdu University of Technology 29 (3), 237–245 (in
Chinese with English abstract).
Zorikin, L.M., Starobinets, I.S., Stadnik, E.V., 1984. Natural Gas
Geochemistry of Oil-Gas Bearing Basins. Mineral Press, Moscow,
USSR. 164 pp. (in Russian).